MICROWAVES AND RADAR

Principles and Applications

’ By the same author :

m Handbook of Electronics
s Electronics and Communication Simplified
m Conventional and Objective Type Questicns and Answers
on Electronics and Communication for Competitions
I +K Electronics Projects for Beginners
I m Digital Electronics
MICROWAVES
AND RADAR
Principles and Applications

kK. MAIN1
F.I.E.T.E.
Senior Scientist
Head of Electronics Division
Laser Science and Technology Centre,
Defence Science Centre (DRDO)
New Delhi -110054

KHANNA PUBLISHERS
2-B, NATH MARKET, NAI SARAK
DELHI-110006.
Phones : 39123 80 ; 722 4179 Fax : 3980311
Published by
Romesh Chander Khanna
for Khanna Publishers,
2-B, Nath Market,
Nai Sarak, Delhi-110006.

All rights reserved.

No part of this publication can be reproduced, stored in a retrieval
system, or transmitted, in any form or by any means, electronic,
mechanical, photocopying, recording, or otherwise without the prior
permission of the Author and Rhanna Publishers.

ISBN No. : 81-7409-129-7

Second Edition : 2001

Price : Rs. 150.00

Corn@‘&& Typeset at:

Softserve Computer Systems
Krishan Nagar,
Delhi-110051
Printed at :
New Nirman Printing Press,
Delhi-l 10006
 Dedicated to my
Wife Renuka
Daughter Nupur and Son Nakul
 Preface
-
Microwaves and Radar Principles and Applications is in-
tended to be used as a text book that comprehensively covers all topics
relevant to the subjcct a t undergraduate and graduate engineering
levels. Each and every topic, be it microwave basics, microwave com-
ponents and devices, microwave measurements or microwave systems
such as Radar, Satellites and so on, has been suitably illustrated with
the help of a large number of illustrations, photographs and solved
problems. The book has been written in a very simple and lucid lan-
guage keeping in focus the syllabi prescribed for the subject in different
institutions and universities. To enhance the utility of the book for the
engineering students preparing for the examination, at the end of each
chapter, there are suitably chosen Review Questions picked from the
question papers of various universities which they should be able to
answer themselves after they have read and understood the contents
of the chapter. Unsolved Problems with Answers enables them assess
the level of their understanding of particular topic alongwith Self
Evaluation Exercises comprising of objective type questions. The
answers to the self evaluation exercise have been given towards the
end of the text.
The text begins with an Introduction to Microwaves covering topics
like microwave frequency bands, microwave signal propagation, ap-
plications of microwaves and so on. Maxwell's Equations, which form
the very basis of propagation of electromagnetic waves, microwaves
being no exception, are covered next in chapter-2. These equations
have been discussed a t length both in terms of mathematical treatment
as well as conceptual interpretation. Transmission Media, Transmis-
sion Lines and Waveguides, are covered in chapter-3. Transmission
Media such as Strip Lines and Microstrips used in case of microwave
integrated circuits (MICs) are also discussed. Passive Microwave Com-
ponents are exhaustively covered in chapter-4. Microwave Tubes and
Serniconductor Microwave Devices are respectively discussed in chap-
ters five and six. In both these chapters, some of the newer varieties
have also been discussed.
Microwave Measurements and Instrumentation constitute the sub-
ject matter of chapter-7 which is devoted to measurement of all impor-
tant microwave parameters like Scattering Parameters, Transmission
and Reflection Parameters, Power, Frequency, Impedance, Noise,
Phase Noise and so on. The test and measuring instruments used to
carry out these measurements including Scalar and Vector Network
Analysers, Spectrum Analysers, Power Meters, Frequency Counters,
Optical Time Domain Reflectometer etc. are also discussed a t length.
 (uiii)

Antennas are discussed next in chapterS. Focus is on antenna fun-

damentals and antenna systems with particular reference to microwave
antenna types. Radar and Satellite systems are respectively covered in
chapter 9 and 10. These two topics have been given exhaustive treatment
beginning with operational fundamentals and then covering types of such
systems and their applications.
Electronic Warfare, which is again very relevant to the study of
microwaves, is discussed in chapter-11 by the same title. This chapter
covers EW fundamentals as well as systems in detail.
The present edition of the book has three major additions. The first
one is in the expansion of chapter-3 on Transmission Media. Based on the
feedback received from a large cross-section of readers, the chapter in-
cludes exhaustive mathematical treatment on waveguides. The second
change is in chapter4 on Microwave Components where some new infor-
mation on microwave components has been included. The third and the
last major addition is inclusion of a new chapter on Microwave Com-
munication Link Design. This new chapter mainly focusses on the opera-
tional and design basics of microwave communication link.
I do hope that the text in the present form will meet the require-
ments of the students doing graduation in electronics and communica-
tion engineering and post graduation in microwaves. I will appreciate
any suggestions and comments from students and teaching faculty
alike so that I can strive to make the text more useful in the editions to
come.

A.K. Maini
 CONTENTS
Chapter-1 :Introduction to Microwaves 1-10
Microwaves
Low Frequencies vs Microwaves
Frequency - Wavelength
Microwave Signal Propagation
Advantages of Microwaves
Applications of Microwaves
Review Questions
Problems
Self Evaluation Exercises

-
Chapter-2 :Maxwell's Equations 11-46
2.1. Vector Algebra
2.2. Faraday's Law
2.3. Biot Savart Law and Ampere's Law
2.4. Displacement Current
2.5. Maxwell's First Equation
2.6. Maxwell's Second Equation
2.7. Maxwell's Third Equation
2.8. Maxwell's Fourth Equation
Solved Problems
Review Questions
Problems
Self Evaluation Exercise

Chapter-3 :Transmission Media, Transmission Lines 47-111

and Waveguides
3.1. Transmission Line Equivalent Circuit
3.2. Sources of Losses in Transmission Lines
3.3. Transmission Line Parameters
3.4. Types of Transmission Lines
3.5. Impedance Matching Using Transmission Lines
3.6. Smith Chart
3.7. Wave Guide
3.8. Waveguide Modes
(ix)
 3.9. Waveguide Parameters
3.10. Plane Wave Propagation in Waveguides
3.11. Transverse Electric Modes in Rectangular
Waveguides - Mathematical Analysis
3.12. Various TE Modes in Rectangular Waveguides
3.13. Transverse Magnetic (TM) Waves in Rectangular
Waveguides - Mathematical Analysis
3.14. Various TM Modes i n Rectangular Waveguides
3.15. Power Loss in Rectangular Waveguides
3.16. Propagation of TE Waves in Circular Waveguides
3.17. Propagation of TM Waves in Circular Waveguides
3.18. Cut-off Wavelength in Circular Waveguides
S.19. Circular Waveguides - Advantages and Disadvantages
3.20. Power Loss in Circular Waveguides
Review Questions
Problems
Self-Evaluation Exercise

Chapter4 :Microwave Components 112-164

4.1. Coupling Probes and Loops 113

4.2. Windows 116
4.3. Tuning Posts and Screws 118
4.4. Waveguide Junctions 119
4.5. Directional Couplers 125
4.6. Isolators and Circulators 130
4.7. Waveguide Flanges 134
4.8. Rotating Joints 136
4.9. Attenuators 136
4.10. Phase Shifters 139
4.11. Cavity Resonators 141
4.12. Wavemeters 147
4.13. Hybrid ring (Rat-race) 149
4.14. Slotted Line 150
4.15. Striplines and Microstripc: 151
4.16. SAW Devices 153
Solved Problems 154
Reviews Questions 160
Problems 160
Self-Evaluation Exercise 161
Chapter-5 :Microwave Tubes 165-219
5.1. Limitations of Gridded Tubes a t High Freqcencies
5.2. Klystron
5.3. Multi-cavity Klystron
5.4. Reflex Klystron
5.5. Magnetron
5.6. Cross Field Amplifier
5.7. Travelling Wave Tube (TWT)
5.8. Backward Wave Osciilator
5.9. Gyrotron
5.10. Comparison of Microwave Tubes
Solved Problems
Review Questions
Problems
Self Evaluation Exercise

Chapter-6 :Semiconductm-Microwave Deirices and 220-291

Integrated Circuits
6.1. High Frequency Limitations of Bipolar Transistors
6.2. Varactor
6.3. Step Recovery Diode
6.4. Parametric Amplifiers
6.5. PIN Diode
6.6. Point Contact Diode
6.7. Schottky Barrier Diode
6.8. Tunnel Diode
6.9. Backward Diode
6.10. Gunn diode
6.11. Transit Time Devices
6.12. IMPATT D i d e
6.13. ,TRAPATTDiode
6.14. BARRITT Diode
6.15. TUNNETT Diode
6.16. MITATT Diode
6.17. Microwave Field Effect Transistors
6.18. MASER
6.19. LASER
6.20. Microwave Integrated Circuits (MICs)
6.21. Materials for making MICs
 (xii)

6.22..Thick Film Hybrid MICs

6.23. Thin Film Hybrid MICs
6.24. Monolithic Microwave Integrated Circuits
Review Questions
Problems
Self-Evaluation Exercise

Chapter-7:Antennas 292-339
7.1. What is an Antenna
7.2. Antenna Reciprocity
7.3. Radiation Mechanism
7.4. Antenna Parameters
7.5. Resonant and Non-Resonant-Antennas
7.6. Electrical and Physical Lengths
7.7. Types of Antennas
7.8. Hertz, Dipole and Marconi Antennas
7.9. Yagi-Uda Antennas
7.10. V and Rhombic Antenna
7.11. Reflector Antenna
7.12. Lens Antenna
7.13. Horn Antenna
7.14. Helical Antenna
7.15. Log-Periodic Antenna
7.16. Phased Array Antenna
7.17. Microstrip Antenna
Review Questions
Problems
Self Evaluation Exercise

Chapter* :Microwave Measurements and

Instrumentation
8.1. Scattering Parameters
8.2. Transmission and Reflection Parameters
8.3. Network Analysers
8.4. Measurement Uncertainty
8.5. Measuring with Scalar Network Analysers
8.6. Measuring with Vector Network Analysers
8.7. Microwave Power Measurement
8.8. Noise Measurement
 (xiii)

8.9. Phase Noise 363

8.10. Time Domain Reflectometry 366
8.11. Spectrum Analysers 367
8.12. Frequency Counters 375
8.13. Frequency Synthesisers and Synthesised Frequency
Generators 381
Review Questions 391
Self Evaluation Exercise 391

Chapter-9 :Radar Fundamentals 393-488

What is Radar?
Basic Radar System
Basic Radar Functions
Accuracy and Resolution
Radar Range Equation
Radar Waveforms
Radar Transmitters
Radar Receivers
Signal Processor
Radar Displays
Radar Classification
CW Radar
FM-CW Radar
Moving Target Indicator (MTI)
Pulse Doppler Radar
True and Apparent Ranges
Tracking Radar
Range Tracking
Velocity Tracking
Pulse Compression Radar
Synthetic Aperture Radar
Over-The-HorizonRadar
Monostatic Radar and Bistatic Radar
Primary and Secondary Surveillance Radar
Laser Radar
Radar Guidance of Missiles
Review Questions
Problems
Self-Evaluation Exercise
Chapter-10:Satellites and Satellite Communications
10.1. Satellite Frequency Bands
10.2. What is a Satellite?
10.3. Satellite Orbits and Trajectcries
10.4. Different Types of Satellite Orbits
10.5. Putting Satellite in the Correct Orbit
10.6. Different Satellites-Orbital Requirements
10.7. Satellite Stabilisation
10.8. Orbital Parr:~r,etms 500
10.9. Eclipses
10.10. Components of Satellite Comunication System
10.11. Multiple Access Systems
10.12. Satellites For Communication Services
10.13. Satellites For Weather Patro; and
Atmospheric Monitoring
10.14. Remote Sensing Applications
10.15. Scientific and Technological Applications
10.16. Military Applications
h v i e w Questions
Problems
Self Evaluation Exercise

Chapter-11:Electronic Warfare 554-587

11.1. Electronic Warfare

11.2. Techniques and Systems - Different Categories
11.3. Electronic Support Measures
11.4. Use of Stealth Tech~ology
11.5. Electronic Countermeasures
- Active and Passive Techniques
11.6. Chaff and Decoys
11.7. Electro-optic Countermeasures
11.8. Electronic Counter Countermeasures
11.9. Current and Future Trends in Electronic Warfare
Review Questions
Self Evaluation Exercise
Chapter-12 :Microwave Communication Link 588-614
Basic Design Considerations
12.1. Microwave Link Design - Basic Considerations 588
12.2. Microwave Path Design a t Millimetric Wavelengths 602
12.3. Analog and Digital Microwave Links 606
12.4. Microwave System Specifications 607
Review Questions 611
Problems 611
Self Evaluation Exercise 612

Answers to Self Evaluation Exercises 615-618

INDEX 619-623
 Introduction to
Microwaves

his introductor~.chapter gives an outline ofwhat the microwaves

T are, different microwave frequency bands and their designa-

tions, modes of propagation of microwaves, advantages that the
microwaves offer particularly in communication related applica-
tions and different areas where microwaves have found use including
commercial, industrial, chemical and research oriented applications.
The topics covered in this chapter include :
1.1. Microwaves
1.2. Low Frequencies us. Microwaves
1.3. Frequency and Wavelength
1.4. Microwave Signal Propagation
1.5. Advantages of Microwaves
1.6. Applications of Microwaves
2 MICROWAVES AND RADAR

1.1. Microwaves
Microwaves refer to the electromagnetic radiation in the frequency
range of several hundred MHz to several hundred GHz. The correspond-
ing wavelengths are in the centimetric millimetric ranges. Microwaves
occupy a region in the electromagnetic spectrum that is bounded by
radio waves on the side of longer wavelengths and infra-red waves on
tbe side of shorter wavelengths as shown in Fig. 1.1.There are infact
no sharp boundaries between these regions except by arbitrary defini-
tion. In the microwave band tco, a further classification is made with
such names as decimetric (for frequencies around several hundreds of

lf
if I
Mtcrowaves
I , 1 , ; '1; (Frequency)

1 I

10" lb2 1 o0 lo-2 (Wavelength)

Fig.1.1

Table 1.1

Frequency Wavelength
Class Range Range

Low Frequency 30 to 300 kHz 10 to 1km

Medium Frequency 300 to 3000 kHz 1000 to 100 m
High Frequency 3 to 30 MHz 100 to 10 m
Very High Frequency 30 to 300 MHz 10 to 1m
Ultra High Frequency 300 to 3000 MHz 100 to 10 cm
Super High Frequency 3 GHz to 30 GHz 10 t o 1cm
Extremely High 30 to 300 GHz 10 to 1m m
Frequency

M H ~ ) ,centimetric (for frequencies around several (GHz) and mil-

limetric (for frequencies beyond 100 GHz) waves. The conventional
method of classifyingelectromagneticwaves is given in Table 1.1.There
is a letter code designation commonly used in case of microwave
frequency bands. These are outlined in Table 1.2.
The historical development of microwaves is only a particular
phase of the gradual evolution of the concept and application of
electromagnetic waves in general. The foundation was laid by James
Clerk Maxwell in 1864 when he put forth a set ofmathematical
INTRODUCTION TO MICROWAVES 3

Table 1.2

Band Frequency Range Wavelength Range

390 - 1550 MHz 76.9 to 19.3 cms.

1.55 - 5.2 GHz 19.3 to 5.77 cms
C 3.9 - 6.2 GHz 7.69 to 4.84 cms
X 5.2 - 10.9 GHz 5.77 to 2.75 cms

equations governingelectromagnetic phenomenon. These equations are

popularly known as Maxwell's equations. Solution of these equations
led Maxwell to predict the existence of electromagnetic waves. Later in
1888, Heinrich Hertz demonstrated experimentally the t r u t h of
Maxwell's equations and thus confirmed the existence of electromag-
netic waves.

1.2. Low Frequencies vs. Microwaves

The electronics technology for microwaves h a s sort of lagged
behind its low frequency counterpart. The knowledge for low frequency
electronics is rich and relatively easy. Infact, what makes designer's or
engineer's life easy a t low frequencies and a bit more difficult a t
microwave frequencies is the size of signal's wavelength. The
wavelength of a low frequency signal is very large as compared to the
size of the components processing them with the result that the phase
difference between the signals a t different test points is extremely
small. More importantly, this extremely small phase difference implies
that a standing wave, which is caused by the interference of forward
signal and its reflection, can not be formed. Consequently, the solutions
of Maxwell's equations are greatly simplified. Tools like Ohm's law and
Network theorems come very handy and make low frequency design
relatively much easier.
On the other hand, a microwave circuit is not easy to analyse. At
n~icrowavefrequencies, the wavelength of the signal is comparable or
sometimes even smaller than the physical dimensions of the processing
components. The finite phase difference between the forward and
reflected waves leads to formation of a standing wave. Also, high

Another important characteristic of nlicrowaves is that they can

propagate through the ionosphere with minimum attenuation. How-
ever, water vapours, rain. ozone, oxygen absorb microwaves of certain
4 MICROWAVES AND RADAR
frequencies. Electromagnetic radiation of frequency up to a few tens of
MHz can not penetrate the ionosphere. These w-.ves are reflected back
towards earth. Microwaves can, however, go Gmugh the ionosphere
and only suffer some rtfiaction. Hence, space bound comn.unications
invariably employ microwaves.
1.3. Frequency-Wavelength
A definite relationship exists between the frequency (f) and the
corresponding wavelength (h) of electromagnetic waves. The product of
these two, i.e. (f) and (h)gives the velocity of propagation of electro-mag-
netic waves and i t equals the velocity of light. This is expressed as
c=f.h
where c = velocity of light (appro*. 3 x 10'') cm/sec)
The value of (c) given above is the velocity of light or electromag-
netic waves in vacuum. We can use the same values in air as well. In a
dielectric medium such as Teflon, Polystyrene or even in water, the
electromagnetic waves are slowed down and this retradation is inver-
sely proportional to square root of the dielectric constant. We know, as
stated above, the product of frequency and wavelength must equal the
velocity. Therefore, the wavelength in a dielectric medium (for a given
frequency) will also be reduced by the square root of the dielectric
constant. That is
Lo
k = ~
where h = wavelength in dielectric medium
ho = wavelength in free space
E,. = relative dielectric constant of the medium
1.4. Microwave Signal Propagation
Microwaves propagate as space waves. This is due to the fact that
the wavelengths at microwave frequencies are too short for any reflec-
tion from ionosphere if there was a sky wave component in the trans-
mitted signal and also at high frequencies, the ground wave component
dies out very close to the transmitter due to the increasing tilt of the
propagating wavefronts caused by diffraction effects.
Space waves travel in line of sight direction and are therefore
limited in range by the curvature of earth. The radio horizon is of course
slightly greater than the optical horizon caused by diffraction effects
around the curvature of earth. The radio horizon can be computed from
(Fig. 1.2) :
~=4m+4=
where B = Distance between transmitter and receiver (in km)
Ht = Height of transmitting antenna above the
surface of earth ( in meters )
Hr = Height of receiving antenna above the surface of earth
(in meters )
INTRODUCTION TO MICROWAVES 5

Fig. 1.2.

Though in case of microwaves, line of sight propagation is almost

implicity assumed, the waves in practice scattered by various objects
or obstacles in the medium by phenomena such as Reflection, Refrac-
tion, Diffraction etc. The reasons why nlicrowaves while propagating in
the atmosphere are subject to reflection, refraction and diffraction
effects just like light waves and unlike radio waves in the very low
frequency (VLF) range are that the wavelengths associated with
microwave frequencies are comparable to or smaller than the physical
dimensions of objects usually interactingwith waves. Laws ofreflection
and refraction that we are all familiar with in refrence to light waves
can be applied to microwaves as well. For instance, reflection of a
microwave by a conducting plane is similar to reflection of light wave
by a metallic mirror with the angle of incidence being equal to the angle
of reflection and the amplitudes of incident and reflected waves being
equal. Analogous to light waves, microwaves undergo diffraction when
they encounter a n obstacle or a n opening which is comparable to or a
bit srnaller than the wavelength.

Fig. 1.3
6 MICROWAVES AND RADAR

Beyond the horizon, propagation of microwaves is feasible by a

phenomenon known as Tropospheric Scatter. Troposphere, a region
which is within 15 km above ground, has certain characteristics that
cause the scatter. Tropospheric scatter is not a sporadic phenomenon.
If the transmitted antenna and receiving antenna beams intersect each
other almost midway between then1in the troposphere region, a fraction
of transmitted energy is forward scattered towards the receiving anten-
na as shown in Fig. 1.3.
Another phenomenon that affects microwave propagation is Duct-
ing also called Super-refraction. Ducting is caused by a temperature
inversion phenomenon due to which in a certain zone above earth's
surface, the temperature instead of decreasing increases with height.
This leads to rapid reduction in refractive index in this region leading
to bendingofmicrowaves. The wave while propagatinggets successively
reflected from the upper layer of the region and the ground surface as
shown in Fig. 1.4. Ducting occurs in a region up to 30 meters from the
surface of the earth.

Fig. 1.4

1.5. Advantages Of Microwaves

High frequencies associated with microwaves offer many ad-
vantges when i t comes to their use for communication related applica-
tions. From the viewpoint of communications, both the high frequencies
or short wavelengths and the consequent high bandwidths available for
use offer many significant features. As an example, the frequency
difference between the S-band (1.55 GHz to 5.2 GHz) and K-band (10.9
to 36 GHz) is about 35 GHz which is more than ten times the frequency
range used for radio com~unicationand TV broadcast put together.
And this range has a great prospect of getting enhanced manifold in
time to come with technological advantages in the field of devices that
can generate these frequencies with reasonable power levels.
Another advantage offered by microwaves, again due to their short
wavelengths, is due to its higher directivity for a given antenna size and
resolvingpower. At microwaves, physically realisable and conveniently
sized antennas can be used to generate highly directive and pencil
beams. At longer wavelengths, antenna size required to produce a
narrow beam becomes prohibitively large. Also, the resolving power,
INTRODUCTION TO MICROWAVES 7

which in essence is the ability of the wave to differentiate one reflecting

object from another, is higher for microwaves. It is because of the above
reasons that microwaves are used invariably in systems like Radars,
Satellite communications.
If we go a step further, waves with still shorter wavelengths such
as IR waves should have a further edge in terms of resolution or
directivity. In earlier days, before the advent of lasers, due to non
monochromaticity of IR radiation, non-coherence of phase and inability
to do manipulations like modulation, amplification etc. i t did not com-
pare favorably with mirowaves. With lasers having arrived in a bigway
and having the properties of monochromaticity, directionality and
coherence which the conventional IR radiation lacked, they have made
a big impact in a large number of applications where microwaves were
a strong candidate such as laser equivalent of a microwave radar,
microwave communication. But due to the fact that lasers are available
only in certain discrete wavelengths and that microwave technology is
very well established in terms of devices, circuitry, techniques etc.,
n~icrowavesstill remain the strongest candidate in communication and
communication related applications.

1.6. Applications Of Microwaves

Microwave technology has found applications in all major areas.
In addition to the use of microwaves in Point to point communication,
Television broadcast services via satellite, Radar systems, Electronic
warfare, they also have a large range of consumer, industrial, biomedi-
cal, chemical and scientific researzh applications.
Microwave ovens, which we are all familiar with, is one of the
popular consumer products where the heating effect of microwaves has
been put to use. Microwave heating also has a large number of in-
dustrial applications including microwave curing of rubber and
materials processing. Agricultural applications include microwave
warming of plants, heating of corn fields, drying of cotton seed, repas-
teurisation of pasteurised milk and so on. Microwaves have also been
used for sensing moisture of wheat, rice etc. fibre orientation in paper,
defect in tyres, hairline cracks in metals, permittivity of dielectrics and
SO on.

Microwaves are used for both diagnostic as well therapeutic pur-

pose. Use of microwaves and detection techniques for monitoring,
imaging and sensing of biological and physiological functions of human
body are very common. Detection of neurology activity in brain, cerebral
blood flow, detection of breast cancer, lung water are some examples.
Treatment of cancer by localised hyperthermia is widely known applica-
tion of microwave energy.
Molecular absorption spectroscopy, Nuclear Magnetic Resonance
(NMR) spectroscopy and chemical synthesis and analysis using
microwave produced plasma a r e the chemical applications of
microwaves.
8 MICROWAVES AND RADAR

Review Questions
1. Briefly outline the main reasons responsible for electromag-
netic waves at nlicrowave frequencies propagating as space
waves in the line of sight and not as Ground waves or Sky
waves .
2. What is the major cause offormation of Troposphere and what
is rts effect on the propagation range of microwaves ?
3 With the help of a diagram, explain what is optical horizon and
how it is different from radio horizon. Which one is larger of
the two ?
4. Briefly describe how high frequencies associated with
microwaves are advantageous as far as their use for com-
munication is concerned ?
5. Briefly outline applications of microwaves other than those
related to comn\unication.

Problems
1. Compute the velocity of propagation of electromagnetic waves
in a medium having a reiative permittivity of 4. (Assume
propagation velocity in free space = 3 x 10' m/s)
[1.5 x 10' m/s]
2. Velocity of propagation of electromagnetic waves in free space
is observed to be 20 percent higher than that in a dielectric
medium having a relative permittivity of ( E ~ Determine
). (EF).
[1.Ul
3. In a microwave communication link using repeater stations,
if the height of antennas a t different repeater stations is 100
meters above ground surface, determine the maximum per-
missible distance between two adjacent repeaters if the line of
sight propagation is to be maintained.
180 kml

SELF EVALUATION EXERCISE

Multichoice Questions
1. An electromagnetic wave having a frequency of 1 GHz is
travelling in a dielectric medium of dielectric constant 2.25.
Its velocity of propagation is
( a )2 xlo8 m / s ( b )3 x lo8 m/s
(c) 4.5 x lo8 m/s ( d ) 1.33 x 10' m/s
2. The wavelength of the travelling electromagnetic wave of Q-1
will be
INTRODUCTION TO MICROWAVES 9

(a)30 cm (b) 20 cm
(c) 45 cm ( d )67.5 cm
3. The wavelength of an electron~agneticwave in free space in
metres is related to its frequency in MHz by
3 x loS
(a) h = -- (b) h = 3 x 10'f
f
f
(d) h = ------
3 x 10'
4. The wavelength (hl) of a n electron~agneticwave travelling in
a dielectric m e h u m ( ~ 1 with
) velocity (C1) is related to its
wavelength (h2) travelling in a dielectric medium ( ~ 2 )with
velocity (C2) by :
hl E2
(a) --=-=- C1
A2 ~1 C2

(c) -hl= - =62-2. C2 (d) -3L=2 - =E2

- C2
h2 EI C1 hl EI C1
5. Millimetric waves correspond to a frequency range of
(a)30 to 300 kHz (b) 30 to 300 GHz
(c) 3 to 30 GHz (d)30 to 300 MHz
6. A frequency range of 1kHz to 10 GHz is spread over
(a)7 decades (b) 7 octaves (c) 20 octaves (d)20 decades
7. When a n electromagnetic wave passes from one medium to
another medium of a different dielectric constant, then one of
the following can happen.
( a )attenuation (b) absorption
( c ) reflection (d) refraction
8. A range of microwave frequencies more easily passed by the
atmosphere than others is known as
(a)atmospheric window ( b )pass band
(c) atmospheric resonance (d)none of these
9. VLF waves are used for some types of services because
(a) they are very reliable
( b ) they penetrate the ionosphere very easily
(c) of the low powers required
(d) of smaller size of transmitting antennas
10. The frequencies corresponding to infrared region of optical
opectrum are
( a ) higher than those corresponding to millimetric waves
10 MICROWAVES AND RADAR
( 6 ) lower than those corresponding to millimetric waves
( c ) in the region of hundreds of gigahertz
(d) higher than those corresponding to ultraviolet region

Fill in the blanks .

The frequency ranges and the corresponding wavelengths of dif-
ferent parts of electromagnetic spectrum are shown in Table 1.3 ahead.
The table has some missing entries. Find the missing information from
the wordslfigures given below :
Infra-red, Near infra-red, Visible, MF, UHF, 100 k m to 10 km,
lo5 to loG,30 Hz to 300 Hz, 3 Hz to 300 Hz, 3 Hz to 30 Hz.
Table 1.3

1 Zd1 Frequency
Range
Wavelength
Range
Band
Designation

1 1 0to loG
~ rn ~ I ELF I
3 300 HZto 3000 HZ 10Gm to 105 ULF
4 3000 HZto 30 kHz (ii) VLF
5 30 kHz to 300 kHz lo4 to lo3 m LF
6 300 kHz to 3000 kHz 103m to 10Zm 1 (iii)

8 30 MHz to 300 MHz 10mtolm VHF

9 300 MHz to 3000 1m to 10cm (iv)
MHz
10 3 GHz to 30 GHz 10 cm to 1cm SHF
11 30 GHz to 300 GHz 1cm to 1mm EHF
12 300 GHz to 3000 1mm to 100pm Very long wave
GHz infra-red
13 3THzto30THz 100pmto10pm Far infra-red
14 30 THz to 300 THz 10 u m to 1u m (v)
1 15 1300THZto 3000 THZ11B m to 100 nm I Near infra-red to 1
I I I ' I ultraviolet I
 Maxwell's Equations

axwell's equations are the gist of electromagnetic theory and

therefore very relevant to the study of any topic in com-
munication engineering, including microwaves. Maxwell's
equations describing interrelationship of electric and mag-
netic fields can be used to explain the behaviour of propagating electro-
magnetic waves. The focus in this chapter is therefore on all important
aspects of Maxwell's equations. The chapter begins with a brief intro-
duction to some of the elementary concepts in vector algebra which are
vital to the understanding of Maxwell's equations. This is followed by
descriptim of these equations in both point as well as integral form.
Each equation is discussed in detail as regards its meaning, interpreta-
tion and applications. Alarge number of solved problems relating to the
use of Maxwell's equations in electromagnetics are given towards the
end of the chapter which concludes in a self evaluation exercise.
The topics covered in this chapter include :
2.1. Vector Algebra 2.7. Maxwell's
- Second
2.2. ~araday's~aw Equation
2.8. Maxwell's Third
2.3. Biot-SavartLaw Equation
2.4. Displacement 2.9. Maxwell's Fourth
Current Equation
2.5. Maxwell's Equa-
-
tions
2.6. Maxwell's First
Equation
12 MICROWAVES AND RADAR

2.1. Vector Algebra

Vector and Scaler
1. A Vector is a quantity having both a magnitude as well as a
direction. Displacement, Velocity. Acceleration, Electric Field Inten-
sity, Electric Current etc. are examples of vector quantities
2. A Scalar is a quantity having a magnitude but no direction.
Length, Time, Mass etc. are scalar quantities.
3. Following are some of the popular laws of vector algebra :
(i) A + B = B + A Commutative Law of
Addition
(ii) A + (B + C) = (A + B) + C Associative Law of Addition
(iii) mA =Am Commutative Law of
Multiplication
(iv) m(nA) = (mn)A Associative Law of Multi-
plication
(v) ( m + n ) A = m A + n A Distributive Law
(vi) m (A + B) = mA + mB Distributive Law
In the above, A, B, C are vector quantities whereas m and n are
scalers.
4. A Unit Vector is a vector having a magnitude of upity. Any
vector quantity (A) can be represented by a unit vector ( a ) in the
direction of vector Oi) multiplied by the magnitude of vector a).
Thatis:A= I A I
The rectangular unit vectors 1 are the unit vectors along the
direction of positive x , y and z axes respectively. Refer to Fig. 2.1.

Fig. 2.1
5. Any vector A can be represented in terms of its component
vectors in the three dimensional rectangular co-ordinante system as :
MAXWELL'S EQUATIONS 13

where Al, A2 andA:j are the magnitudes of component vectors along

axes respectively.
x, y, z

The magnitude of vector A is given by :

(A I dA12 + ~2~ + ~ 2 ~
As a special case, position vector (r) from origin to the point
(2, y, z) is given by :

r = (xf+yftzk) with IrI =

6. When to each point (x, y, z), of a region (R) in space, there
corresponds a number or a scaler $ (x, y , z), then ($) is called a scaler
function of position or in other words, a scaler field has been defined in
region (R).A scalar field that is inde-
pendent of time is called a stationary
or a steady state scaler field.
When to each point (x, y, z) of a
region (R)in space, there corresponds
a vector V (x, y, z), then V is called a
vector function of position or in other
words, a vector field is defined in
region
pendent (R). A vector
of time is called
fieldstationary
that is inde-
or L LA

a steady state vector field. Fig. 2.2.

Dot-Product and Cross-Product
The Dot product or Scalar product of two vectors A and B, denoted
by A.B (It should be read as A dot B and not as A multiplied by B), is
defined as the product of the magnitudes ofA and B and cosine of the
angle (0) between the two. Mathematically
A.B =AB cos 8 where 0 I0 IIT (Fig. 2.2).
The dot product of two vectors is always a scalar. Following are
some of the useful mathematical identities with reference to dot
product.
(i) A.B = BA (Commutative law for Dot products)
(ii) A. (B + C) =A.B +A.C (Distributive law)
(iii) m (A.B) = (mA). B = A . (mB) = (A . B) ~n(m is a scalar)
f i A A & Ah h
z.z=j.j=RR=l,zj=j.k=k.E=~
(iv) (:Sand R are unit vectors along x, y and z axes respectively)
(u) If vectors A and B are not null vectors and are such that
A. B = 0, then vector A and B are n~utuallyperpendicular.
(vi) If vectors A and B are expressed in terms of their component
vectors as
-
A =A1 ~+Az!+A~& and B B1 E+ B~!+ B~ &,
14 MICROWAVES AND RADAR

B . B = B ~ = B ~ ' + B ~ ~ + B ~ ~
and A.B=B.A=A1Bi+AzBz+A&j
2. The Cross Product of vectors A and B, denoted by A x B (read
as A cross B ) , is also a vector and that is why cross product is also
referred to as a Vector Product. The magnitude of this vector is
(AB Sin a). The direction of the vector C = A x B is perpendicular to the
plane carrying vectors A and B. See Fig. 2.3.
Mathematically, A x B = (AB sin 0) 2 where (2) is a unit vector
indicating the direction of vector C ( = A x B ) and 0 < 0 Ix.
Following are some of the useful mathematical identities with
reference to cross-product.
(i) A x B = - B x A
(Commutative law for Cross-product is not valid)
(ii) A x ( B + C ) = A x B + A x C
(Distributive law)
(iii) m ( A x B ) = ( m A ) x B = A x ( m B ) = ( A x B ) m
( m is a scalar)

Fig. 2.3
A A 4 ..\
(iu) z x z = j x j = R x k = ~ a n d E x f = $ , ~ ~ R = E a n d R x E = f
( v ) IfA =A1 ;+A$+A&, B = B1 E+ B$+ B3 k, then

(vi) If vectors A and B are such that A x B = 0, the A and B are

parallel provided they are not null vectors.
(vii) The magnitude of the cross-product of vectors A and B is the
same as the area of the parallelogram with sides A and B.
MAXWELL'S EQUATIONS 15
3. Following are some of the useful expressions involving Dot and
Cross products of three vectors.
( i ) (A. B ) x C # A x (B.C)
(ii) A. ( B x C ) = B. (C x A ) = C. (A x B )
I t equals the volume of a parallelopiped havingA, B and C as edges
or the negative of this volume depending upon whetherA, B and C form
a right handed system or not.
If A =AI LA~S+A~R
and
A1 A2 A3
A. ( B x C ) = B i B2 B3
c1 C2 c3
(iii) A x (B x C ) # (Ax B ) x C
(iv) A x ( B x C ) = (A. C ) B = (A. B ) C
(v) ( A x B ) x C = ( A . C ) B - ( B . C ) A
The product A . ( B x C ) is also referred to a s scalar triple product.
The productA x ( B x C )is also referred to as vector triple product.
Vector Differentiation
1. If vectors A, B and C are differentiable vector functions of a
scalar ( u )and if ($) is a differentiable scalar function of ( u ) ,the following
mathematical identities are valid.
d dA d B
(i) - ( A + B ) = - + -
du d u du
d dB dA B
(ii) -(A.B)=A.-+-.
du du du
d dB dA
(iii) - ( A x B ) = A x - + - X B
du du d u
d d'4 d$
(iv) - ( $ A ) = $ - + - A
du du du

2. Partial derivatives of "ector f;nctions are defined when vector

A depends upon more than one scalar variables. If x , y and z are the
scalar variables on which vector A depends, then A is written as
A(x, y, z ) and the partial derivatives ofA with respect to r,y and z are
respectively expressed as :
16 MICROWAVES AND W A R

dA
- - A(x,y+Ay,z)-A(x,y,z)
- lim
Ay -t 0 AY
and
&4
-- - lim A ( x , y , z + 1 2 ) - A (x,y,z)
az h + 0 Az
3. Higher order partial derivatives are also defined as in case of
calculus

(z z t ) " A =- -
2
az
aazq
\ /
4 . If A and B are vectors depending upon scalar variables x , y, z
then

(ii) - ( A
a aB ? A
a x x B ) = A -a+x- XaBx
a2
(iii) -( A . B ) = -a1 -T a ( A . B )
ayax JY

5 . I f A =A1 ; + A ~ ; + A $ , then dA =&I;+ dA2;t d~?f

6.d(A.B)=A.dB+dA.B
7. d(AxBj=AxdB+dAxB
Gradient-Divergence-Curl
1. The vector differential operator Del (denoted by V ) is defined
as

and Curl of vectors.

MAXWELL'S EQUATIONS 17

2. If $ (x,y, z ) is a scalar field that is defined and differentiable at

each point (x, y, z) in a certain region of space, then the gradient of (9)
expressed as Grade + or V $ is given by

Although (I$) is a scalar field, (V $) is always a vector field.

3. If V (x, y, z) is a vector field that is defined and differentiable
at each point (x, y, z) in a certain region of space, then the Divergence
of the vector field written as Div V or V . V is given by

4. If V (x, y, z) is a differentiable vector field, then the Curl V or

Rotation of V written as o r V x V is given by

$Ir
R
-a
-
ax aay -
aaz
Vl v2 v3

5. If A and B are differentiable vector functions and (I$) and (v)

are differentiable scalar functions of position (x, y, z) the
following mathematical identities are valid.
(i) V ( $ + \ y ) = V $ + V y
(ii) V.(A+B)=V.A+V.B
(iii) V x ( A + B ) = V x A + V x B
(iv) V.(I$A)=(V$).A+$(V.A)
(v) V x ( $ A ) = ( V $ ) x A + $ ( V x A )
(vi) V.(AxB)=B.(VxA)-A.(VxB)
(vii) V x ( A x B ) = ( B . V ) A - B ( V . A ) - ( A V ) B + A ( V . B )
(viii) V ( A . B ) = ( B . V ) A + ( A . V ) B + B x ( V x A ) + A x ( V x B )
18 MICROWAVES AND RADAR

ay2 a z 1
where v2= a [s-a +5
+ is Laplacian op zator

(x) v x (V +) = 0 (CUI-1of the eadient of a scalar field is zero)

(xi) V . (V x A) = 0 (Divergence of curl of a vector field is zero)

In expressions (ix) and (x), it is assumed that $ has a continuous

second order partial derivative. In expressions (xi) and (xii), it is
assumed that (A) has continuous second order partial derivative.
Vector Integration
1.An integral which is to be evaluated along a curve is called a
Line Integral. Such an integral can be defined in terms of limits of sums.
~f a vector A (2, y, z) = ( ~ : +1A ~ J ~ + A S$1 is a vector function of
position defined and continuous along a curve C, then the integral of
the tangential component ofA along C from Pi to P2 written as :

It is an example of line integral.

2. If A = V $ everywhere in a region R of space, defined by
+
a1 5 x 5 a2, b l iS y 5 b2 and cl 5 z 5 c2 where (x,y, z ) is single valued
and has continuous derivatives in R, then

(i) jP2
P1
A. dr is independent of path C in region R joining points
Pi and P2.

f A. dr = 0 around any closed curve C in the region R.

In such a case, A is called a conservative vector field and $ is its

scalar potential.
In simple words, a vector field A is said to be conservative if and
only if V xA = 0 or equivalently A = V +. In such a case,
A. dr = (A1 dx +A2 dy +A3 dz) = d$.
3. Refer to Fig. 2.4. The integral

is an exampleof a surfaceintegral. In the aboveintegral,cESis the vector

whose magnitude is the differential area ds and whose direction is that
of n. The above surface integral is called the flux of A over S. Other
surface integrals are
MAXWELL'S EQUATIONS 19

The notation j j or
S
$$ is used to indicate integration over a

closed surface S

Fig. 2.4
4. Consider a closed surface S in space enclosing a volume
Then I I A dV and I I I $ dV are examples of volume integrals.
v v
Some Useful Theorems in Vector Algebra
1. Divergence theorem :Divergence theorem states that if (V)
is the volume bounded by a closed surface S, and if (A) is a vector
function of position with continuous derivatives, then

where n is the positive normal to S.

2. Stokes' theorem :Stokes' theorem states that if S is an open
two sided surface bounded by a closed non-intersectingcurve C, then if
A has continuous derivatives.
m

Curve C is traversed in the positive direction. The direction of C

is called positive if an observer while traversing the boundary of S in
this direction with the head pointing in the direction of the positive
normal to S, has the surface on his left. You would notice that while
Stokes' theorem converts a line integral into a surface integral. Diver-
gence theorem converts a surface integral into a volume integral, the
20 MICROWAVES AND RADAR
former involving curl operation and the
latter involving divergence operation.
2.2. Faraday's Law 4
A current carrying conductor
produces a magnetic field. In 1831,
Faraday found that a magnetic field
can produce a current in closed circuit.
the converse is also true provided the
magnetic flux linking the circuit is
changing and there is a rate of change
of flux. Refer to Fig. 2.5 in which a
magnetic field with flux density B is
decreasing
normal to theiplane
n magnitude If B is
of the loop.and is $e'OOp B(Decreas1ng)

directed upwards, then a current I Fig. 2.5.

flows in the conductor as shown in the figure. However, if B is is
increasing in magnitude and directed upwards, the direction of induced
current is opposite to that shown in Fig. 2.5.
Faraday's Law gives a quantitative relation between the e.m.f.
induced in a closed circuit and the changing magnetic field that
produces it. According to this law "The e.m.f. induced in a closed circuit
is equal to the time rate of change of magnetic flux linking the circuit".
Mathematically, it is expressed as :

where e = Induced e.m.f.

4 = Flux
t = Time
Negative sign here indicates that the direction of induced e.m.f.
opposes the cause that produces it. Now, the cause that produces the
induced e.m.f. is the changing flux linking the conductor. The negative
sign would thus imply that the direction of the induced e.m.f. is such
that the flux produced by the current due to the induced e.m.f. opposes
the flux that caused the e.m.f. This statement defining the polarity of
induced e.m.f. is often referred to as the Lenz low. The above expression
(Equation 2.1)is for a one t u n conductor.If the conductor has (N) turns,
the expression gets modified to :

Since the total flux through a circuit can be expressed as the

integral of the normal component of the flux density, B over the surface
bounded by the circuit, (4) and ( B )are mathematically related by :

$4 S
B.& (2.3)
MAXWELL'S EQUATIONS 21

Substituting the value of (4) from Eq. (2.31, in Eq. (2.11, we get

(2.4)

where e = Induced e.m.f.

B = Flux density
ds = Surface area
and t = Time
Eq. (2.4)may be applied to circuits in motion with respect to the
observer, with flux density B either constant or time varying, provided
that the motion of the boundary of the surface in Eq. (2.4)coincides at
every instant with the motion of the parts of the physical circuit. When
the loop or closed circuit is stationary or fixed, Eq. (2.4) reduces to

Also, e= f E. dl (2.6)
where E = the electric field produced by changing magnetic fields.
From Eqs. (2.5) and (2.6),we have

where E = Electric field

e = Induced e.m.f.
dl = Element of path
ds = element of area
and B = Flux density .
The expression

is nothing but Maxwell's second equation written in integral form. If

the line integral of ( E ) in the above equation is replaced by its cor-
responding surface integral, we get :

Since ds in Eq. (2.8)applies to any surface element, it is arbitrary

and, therefore, Eq. 2.8 leads, to :
22 MICROWAVES AND RADAR
which is expression of Maxwell's second equation in differential form.
Maxwell's equations alongwith their interpretation are described in
detail in latter part of the chapter :
2.3. Biot-Savart Law and Ampere's Law
PPcordingto the Biot-Savart law, a differential magnetic field (dH)
resdting from a differential current (Idl)is inversely proportional to
the square of the distance and is independent of the surrounding
mdum.

1'b direction of (R) must be from the current element to the point
at which (dm is to be determined. In the integral form of Biot-Savart
lew (Eq.2.18),a closed line integral is required to ensure that all current
elements are included.
According to Ampere's law, the line integral of tangential com-
ponent of magnetac field strength around a closed path is equal to the
current enclosed by the path.

Mathematically,
I
2.4. Displacement Current
H . dl = I (2.19)

The concept of displaceiient current is very vital to the under-

standing of Maxwell's first equation. This concept can be illustrated by
taking a parallel R-Cnetwork as shown
in Fig. 2.6 with a voltage (V)applied
across the parallel combination.
The expressions for current (Ii)
11
through (23)and (I2) through (C)can be

and
cdv
I 2 =-
dt
Fig. 2.6. The current, in practice, does not
flow through the capacitor. However, the current that flows out of one
electrode of the capacitor equals the current that flows into the other
electrode. The net effect is as if there is a current flowing through the
path containing the capacitor. On the other hand, the current (II)
flowing through the resistor actually flows through the resistance
element.
MAXWELL'S EQUATIONS 23

In view of the above, current flowing through the resistor is known

as Conduction current, while the current through the capacitor is
commonly known as Displacementcurrent. The electric field inside each
element equals the voltage V across the element divided by its length,
d. That is

We know that the current density, Ji inside the resistor is given

where A = cross-sectional area

o = conductivity of the resistor materie

Also, capacitance of a parallel plate capacitor is

where d = spacing between the plates.

From Eq. (2.191,

Therefore, the current density, J2 inside the capacitor is given by

If (D)is the electric flux density,

then D =E .
dD
which gives J2 = - (2.25)
dt
Thus (J1)represents the conduction current density denoted in
eneral by (J)and (Jz)is the displacement current density given by
with D = EE and J = oE. The total current density is the sum of
the conduction and displacement current densities.
It may be remembered that current density J ; electric field
intensity E and electric flux density D, are space vectors having the
same direction in an isotropic medium.
24 MICROWAVES AND RADAR

2.5. Maxwell's Equations

Maxwell's equations are a set of four equations relating time
varying electric and magnetic fields. These equations form the basis of
electromagnetictheory and particularly the propagation of electromag-
netic waves.
Maxwell's equations in differential form are as follows :

4. V . B = O
Maxwell's equations in integral form are as follows :

The first two Maxwell's equations in the differential form are

equivalent to their correspondingintegral forms by appliction of Stoke's
theorem. Divergence theorem can be used to equate the differential and
integral forms of last two equations.
Each of the above mentioned equations is described in detail in the
following paragraphs.

2.6. Maxwell's First Equation

Ampere's law forms the basis for Maxwell's first equation.
Ampere's law states that 'Line integral of H around the closed loop
contour is equal to the current enclosed'. The current referred to here is
the total current equal to the sum of conduction and displacement
currents. Mathematically

Substitutifig for Jcond and J d i s ,we get

MAXWELL'S EQUATIONS 25

where J refers to conduction current density. This is the integral form

of Maxwell's first equation. By applying Stoke's theorem to Eq. (2.28)

Maxwell's first equation stades that the magneto-motive force

around a closed path equals the conduction current plus the time
derivative of electric dwplacement through any surface bounded by the
path. In other words, magnetic voltage around a closed path equals the
electric current through the path. Maxwell's first equation is an expres-
sion of Ampere's law for time arying fields.
Maxwell's first equation can be proved as follows (Refer to Fig. 2.7):
*
H.dl =It
For this loop

Now, f H.dl= j
ab
H.dl+ J-
bc
H.dl+
cd
H.dl +
da
Hdl

Fig. 2.7
26 MICROWAVES AND RADAR

and Hda=-[~z+dx~]
aHz

So that

f abcd H.dl = (- Hx) (Ax) +Hz& +

Similarly for liops perpe&dicularto X and-2,we have

Jcx +- soy aHz

and
Multiplying ~~s.'(2.30),
(2.3i) A d (2.32)by;, i and k respectively
and adding, we get

and

2.7. Maxwell's Second Equation

Maxwell's second equation is an expression of Faraday's law
discussed in detail a little while ago. The differential form of this
equation is written as:

The same equation when written in the integral form is :

r r
MAXWELL'S EQUATIONS 27

According to Maxwell's second equation the electromotive force

around a closed path is equal to the time derivative of magnetic
displacement through any surface bounded by the path. In other words,
electric poteniial around a closed path is equal to the magnetic current
through the path. Maxwell's second equation can be proved as
follows :
Considering a rectangular loop in XZ plane as shown Fig. 2.8. The
electric field along X-axis is Ex and those along Z-axis and Y-axis are
Ez and Ey respectively.

Fig. 2.8
E . dl = - at * for the loop in XZ plane

-
-- (br d l )
at
where dA=k.&
If (Hy)is the magnetic field perpendicular to the plane of the wave,

then, f E . d l = - p m at
. 4~& (2.37)

NOW $
abed
E . d l = j ab E . d l + j bc E . d l + j cd E . d l + j da E.dl (2.38)

E along TEG = - Ex along 62;= Ez

along Z l = E x + -a E x . A z
az
28 MICROWAVES AND RADAR

andalong&=-

Substituting the values of E in Eq. (2.38)

Comparing Eq. (2.39)and Eq. (2.37)

Similarly taking loops inXY and YZ planes, we get

Multiplying Eq. 2. by unit vector Eq. 2.40by unit vectorJ"and

P f
l$qh 2.41 y unit vector and then adding the three equations, we get
(t, J and a r e unit vector alongX, Y, and Z axes respectively.

aH aB
whichgives - p - =
atv x E = - - -
at
2.8. Maxwell's Third Equation.
Maxwell's third equation is a mathematical interpretation of
Gauss law which states that total electric flux through any closed
- -
surface surrounding charges is equal to the total charge enclosed.

That is, J D . ds J p, . dv
= (2.44)
MAXWELL'S EQUATIONS 29

Applying Divergence theorem to Eq. 2.44,

V.D=p (2.45)
Here the surface integration is carried out over a closed surface
and the volume integration throughout the region enclosed. If a volume
is devoid of any charge, then the electric flux over the surface of the
volume is always zero even though the volume may be in an electric
field. In such a case, the net flux is zero since the inward flux is equal
to the outward flux.
From the definition of Divergence operator,

So that

where dV = an elemental volume = dx dy dz

Therefore,

Now,

Substituting for I I D . I , we get

Since the volume chosen is arbitrary, the above relation holds for
any volume whatsoever. Moreover as the volume consideredis reducible
to an elemental volume, this becomes a point relation and the in-
tegrands can be equalised.
That is, VD=p (2.49)
2.9. Maxwell's Fourth Equation
Maxwell's fourth equation signifies that there cannot be isolated
magnetic poles on which magnetic lines can terminate and that mag-
netic lines are continuous.
Mathematically,
V.B=O (2.50)
Applying Divergence theorem to Eq.2.50,
30 MICROWAVES AND RADAR
SOLVED PROBLEMS
Problem 2.1 :Let the two regions having relative permeabilities
of 3 and 5 be defined by x < 0 and x > 0 respectively. If the magnetic f i eld
intenstty ( H i )in region-1 is given by :

then determine the magnitude of magnetic field intensity (Hz) in

region-2.
Solution : H i = (4ax + 3ay - 6 & ) A / m
Therefore B I = po k,Hi
= p3 (12 c4, + gay - 18 a,) A / m
Since normal component of (B)is continuous across the interface,

[ Kg
Therefore, B2 = p~ 1 % +~9~- aY - 18 (E) az]

Problem 2.2 : Refer to the interface of two dielectric regions as

shown i n Fig. 2.9. For pi = 3 anc. p z = 5, find expressions for (Bz)and
(Hz)given that

Fig. 2.9
Solution : B1= 2ax +
Choosing the unit normal an = f*} we get
MAXWELL'S EQUATIONS 31

Therefore, Bnl = = 0.59 +0 . 5 ~ ~

Also, Bn2 = Bn, = 0.5% +0 . 5 ~ ~

The tangential component of Bl is given by
Bt, = Bi - Bn, = (2ax + 9 )- (0.5 ay + 0 . 5 ~ ~ )
= 2ax + 0.59 -0 . 5 ~ ~

This gives Ht, = ax + 3%- -az

= Htz
or St, = PO prz Htz = 3.3 ax + 0 . 8 -
~ 0~ . 8 ~ ~
NOW B2 = Bnz + Btz = (0.5 9 + 0.5 a,) + ( 3 . 3 +~0~ . 8 -~ 0.8
~ a,)
=( 3 . 3 +~ 1.3ay
~ -0 . 3 ~ ~ )

Problem 2.3 : Determine expressions for electric flux density D,

magnetic flux density B and magnetic field intensity H in free spacegiven
that
E =Em sin ( k t - $2) cty
Solution : The electric flux density, D is given by
D = E & = E o E sin(ot
~ - pz)%
Magnetic flux density can be determined by using Maxwell's
equation.

ax a, a,
Now VxE=
a
- a
- -a
ax a~ az
O Em Sin ( a t - p ~ ) 0
which gives
32 MICROWAVES AND RADAR

The constant of integration which would be a static field has been

ignored.
From B = MH

Problem 2.4 :Sketch the electric f eld intensity E and magnetic

field intensity H at t = 0 for the expressions i n Problem 3.3. Wh,at is the
direction ofpropagation of the wave ?
Solution : E = E m Sin (wt - pz) a,
-
Hz----- Sin (at- pz) ax
@PO

Em Sin fi
~=(0110 )
The E and H are sketched in Fig. 2.10.
a,

Fig. 2.10.
The wave is propagating in the z-direction.
MAXWELL'S EQUATIONS 33
Problem 2.5 : The magnetic field intensity H of a certainpropagat-
ing electromagnetic wave i n free space is given by
H = Hm & (at + 0 4 a x
Find electric field intensity vector E.
Solution : H = H , 2 (Ot + pZ)a,
From a Maxwell equation

where D = Electric flux density

Problem 2.6 : It is given that

D = Dm sin ( a t + pz) ax
Using Maxwell's equations, prove that
-a'oDm -
P= s i n ( a t + P z ) a y =- sin ( a t + pz) ay
p OEO
Solution :The relevant Maxwell equation is

D Dm
E = - = -sin
EO EO
(at + pz) a,
ax 9 az

VXE=
a
-
ax
aay aaz
Dm
EO sin (a+ pz) 0 0
34 MICROWAVES AND RADAR

Integrating both sides,

-
--@sin(ot+&)a,=-- OWDm sin (at + @) a, (substituting
mo - P
for EO from a/& l/qpo~o)
The problemcan also be solved by using another Maxwell equation
aD
VxH=-
at
D = D msin ( o t + @) ax

This assumes that H vector is in y-direction.

Integrating both sides,

H=-*sin(&+&)a.y
P

Problem 2.7 : Determine the amplitude of the displacement currnt

density in the air space within a large power transformer where
H = lo6 cos (377 t + g.2566 x loT6z) ay Amplmeter.
Solution :The displacement current density
ing to the Maxwell's equation by (V x H) as
density ( J )in the air space would be zero.
ax a, a2

VXH=
a
- a
- a
-
ax aY az
0 lo6 cos (377t + 1.2566 x lo4z) 0
MAXWELL'S EQUATIONS 35

a
=-
az (lo6 cos (377t t 1.2566 x 10%)) ax

= - 1.2566 x lo4 x lo6 sin (377t + 1.2566 x lo4z) ax

= - 1.2566 sin (377t + 1.2566 x 10%)ax
Amplitude of displacement current density = 1.2566 A/m2

Problem 2.8 :Find the amplitude of displacement current density

i n the air near a n antenna radiating FM signal at a point where the
electric field strength is E = 80 cos (6.277 x lo8 t - 2.092 y ) az V/m.
Solution :
E = 80 cos (6.277 x 10' t - 2.092 y) a2

-a 3 a
-
Therefore, V x E =
ax -
ay az
0 0 80 cos (6.277 x lost - 2.092 y )
Electric flux density D = EO E
= 8.85 x 10-l2 x 80 cos (6.277 x 10% ax
- 2.092~)

10-l2 cos (6.277 x lost - 2.092~)ax

= 708 x

Displacement current density = -

aD
at
2D
Now -= - 708 x 10-l2 x 6.277 x lossin (6.277 x 10%- 2.092 y ) ax
at
= - 0.444 sin (6.277 x 10' t - 2.092 y) ax
Therefore, amplitude of displacement current density = 0.0444
wrn2

Problem 2.9 : I n a region where o = 0, cr = 1, = 1, magnetic

vector potential and scalar electric potential are respectively given by :
A =(1fl y cos 3 x 10% cos z) az Wb/m
\
and V= 3 x lo5 sin 3 x 108t sin z volts.
Determine electric field intensity (E) and magnetic field intensity
(H).
Solution :Magnetic flux density (B) is related to magnetic vector
potential (A) by :
B=VxA
36 MICROWAVES AND RADAR

= lo3 a, cos 3 x lost cos z

1
y cos 3 x 10' t cos z a,

= 796 ax cos 3 x lost cos z

Electric field intensity can be computed from
E = - V V - - aL4
at

= - 3 x 1 0sin~3 x~lost cos z

Therefore,
~=-3x10~sin3xl0~tsinz-3x10~~sin3xl0~tcosz
+3~10~~sin3~10~tcosz
=-3x lo5 s i n 3 x lost sinz

Problem 2.10 : Find the amplitude of displacement current den-

sity inside a capacitor where D = 3 x
c/m2 and E~= 100.
Solution : Displacement current density = -
aD
at
= 3 x lod7 x 6 x lo7 cos

Amplitude = 18 A/m2

Problem 2.11 : I n free space, B = Bm d(&+ &) q

Show that :
E = - de "m& + " ' ax
P
MAXWELL'S EQUATIONS

Solution : B = Bm J (Ot PZ)~y

+

Now B = PO H which gives

According to Maxwell's equation,

Problem 2.12 : Given that in fiee space,

and H = H , dlogt+ay A / m
Find $ and Hm ($ > 0 )
Solution : The expressions given for E and H vectors are repre-
sentative of a plane wave.
38 MICROWAVES AND RADAR
9
Also, fiom given expressions, 61 = 10

A80 -=
H
*=
Therefore, P = - 1c9 = -
10
3 x loe 3
rad/m

120n fi for frrc space

3 g n ofHm can be determinedby applying the Maxwell's equation,

=jg 4on J (lo9t + IM ny (1)

and
aB aH
--=-~-=-j109w~,ReJ(
-
logt + p)
9 (2)
at at
Comparing (1) and (2) shows that Hmmust be negative
Hence Hm - 1/3 A/m.

Problem 2.13 : In a honeogeneous medium for which cr = 49 and

= 1, the electric f eld strength vector (E) and magnetic flux density
vector (B) are expressed as
E = 2 0 n @M-BZ)ax V / m
and B = ~ H ~ $ ( ~ Yes& + @ ) ~
Determine (a)and (Hn) if the signal wavelength is 1.8 m
Solution : The velocity of propagation of the wave in the medium
having relative permittivity of 49 is given by :

Wavelength = 1.8 m

Therefore, o = 2n f = 2n 0.43 lo8= 1.6 x

1.8
MAXWELL'S EQUATIONS 39

Therefore, Hm = -
Em = - -*
2ox - 1.17 Urn
17n 17n
Sign of (Hm) can be determined by evaluating the Maxwell's
equation,

Comparing (1)and (21, sign of (Hm) must be positive.

Therefore, Hm = 1.17 Urn.

Problem 2.14 : Find the amplitude of displacement current den-

sity inside a typical metallic conductor having a = 5 x 10~rnho/m,
~r = 1, f = l k H z for J = lo8 sin (ot- 4442) ax ~ / m '
Solution :f = 1 kHz
o = 2nf = 27t x lo3 = 6280 rad/s
which gives J = lo7 sin (6280t - 4442) ax A/m2
Now J = aE which gives

E = - = - lo7 sin (6280t - 444.) ax

0 5 lo7
= 0.2 sin (6280t - 4442) ax
= 8.85 x 10-l2 x 0.2 sin (6280t - 444%)ax
D =E~GE
a0
Displacement current density = -
at
= 1.77 x 10-l2x 6280 cos (6280t - 4442) ax
= 11.11 x 1 0 - l ~wm2.
40 MICROWAVES AND RADAR
Review Questions
1. Write the set of Maxwell's equations both in the differential
and integral forms. Briefly describe the meaning of each of
these equations with particular reference to laws/theorems of
electromagnetics they belong to.
2. Define 'Divergence' and 'Curl' operators. What is the sig-
nificance of Divergence theorem and Stoke's theorem in the
expression of Maxwell's equations.
3. Write down the expressions for those Maxwell's equations
which represent.
(i) Ampere's law for time varying fields
(ii) Faraday's laws
(iii) Gauss theorem
4. Differentiate between Conduction current density and Dis-
placement current density.Write down the relevant Maxwell's
equation involving the two. Rewrite the same expression for
free space. Justify your answer.
5. Read the following statements. Briefly describe the meaning
of each of them with particular reference to the Maxwell's
equations they represent.
(a) Magneto-motive force around a closed path equals the
conduction current plus the time derivative of electric
displacement through any surface bounded by the path.
(b) Electromotive forcgraround a-closed path is equal to the
time derivative of magnetic displacement through any
surface bounded by the path.
(c) Monopole is non-existent.
(d) Total electricfluxthrough any closed surface surrounding
charges equals the total charge enclosed.

Problems
1. An electromagnetic wave characterised by :
E = 20 x: $' - (4I3bIaz V/m
and H = $' I d - W3b1 wm
is propagatingin a homogeneous nonconductingregion having
CLI.= 1. Determine (i) o (ii) g (iii) direction of propagation of
wave.
0' rad/s (ii) 36 (iii) y-direction]
[ (i) 6.66 x 1
2. The magnetic field intensity of a uniform plane wave in air is
MAXWELL'S EQUATIONS 41

Determine the amplitude and direction of electric field inten-

sity E.
[9425 V/m, x-direction]
A
3. A uniform plane is travelling in the ax direction in free space.
The electric field intensity has a maximum amplitude of (200
&+ 400;~) Vlm a t a point P (10, 30, - 40) a t time t = 0. Write
expression for E(x, y, z, t). With the help of one or more
Maxwell's equations, determine H (x, y, 2, t)

4.
E = (200 & + 400 $4 Acos (6 x l0%t - m)V/m
H = (- 1.06 & + 0.53 a,)cos (6x 10% - m ) A/m 1
A wave is propagating in a lossless dielectric. It is expressed
by:
E = 500 cos (107t- 0.0672) 2,V/m
and H = 1.1cos (107t - 0.0672) & A/m
If the wave is traveling at 1.5 x 10' m/s, determine relative
permeability (%) and relatie permittivity ( E ~ of
) the medium.

SELF EVALUATION EXERCISE

Multichoice Questions
1. Ohm's law when applied to electromagnetic phenomenon gives
the following expression :
(a)J=oE (b)V=IR (c) o = J E (d) E = J o
2. Four fundamental equations of electromagnetics are grouped
under
(a)Fleming's laws (b) Faraday's laws
(c) Lorentz equations (d) Maxwell's equations
3. The electric flux density (D)
and the electric field intensity (El
are interrelated by :
42 MICROWAVES AND RADAR
4. "The total flux of a closed surface is equal to the net charge
within the surface". This statement is an expression of
(a)Divergence theorem (b) Gauss law
(c) Faraday's laws (d) Maxwell's equations
5. The electric field intensity ( E ) and electric potential (V)are
interrelated by :
(a)E = - Divergence of V (b) E = Divergence of V
(c) E = - Gradient of V (d) E = Gradient of V
6. The electric field intensity at a point (1,2,2,) m in Cartesian
co-ordinates is (assuming ax, ay and az to be the unit vectors
along x, y and z axes respectively).

7. Laplace equation, which is an extension of Poisson's equation

for regions that have uniform ( E ) and are charge free, when
expressed in cartesian co-ordinates is
a2v a2v+--
(a) -+-
a2v av av av
(b) -+-+-=O
ax2 ay2 az2 - O ax ay az
(d) none of these

[ 5)
8. The Maxwell equation V x H = Jc + - for free space becomes

(a) V x H = O (b)VxH=Jc
aD
(c) V x H = -
at (dl none of these
aB
9. Maxwell equation V x E = - -is an expression of
at
(a) Gauss law (b) Faraday's law
(c) Ampere's law (d) Biot-Savart law
10. Maxwell's equations of electromagnetia are valid under all
conditions except one and that is
(a) these do not apply to isotropic media
( b ) these do not apply to non-homogeneous media
(c) these do not apply to media which move with respect to
system of co-ordinates
(d) these do not apply to non-linear media.
MAXWELL'S EQUATIONS 43

11. One of the followingis the correct expression for one of the four
fundamental equations of electromegnetics.

JD
(c)VxB=O (d)V.H=J+-
at
12. According to wave equation for electric field (E),
J'E
(a) V. E = p~eo- (b)VxE=k~o-
at2 at2

13. According to wave equation for magnetic field (H),

a2H aH
(a)v2H = p , , ~- (b)VxH=k~,,-
at2 at

14. Plane electromagneticwaves travel in vacuum with a velocity

given by :
7

15. For a plane electromagnetic wave travelling in a linear,

isotropic,homogenous and stationary medium,
(a) E and H are in the same plane
(b) E and H are mutually perpendicular
(c) E x H = O
(d) none of these
16. One of the four Maxwell's equations can be interpreted to
imply that magnetic lines are continuous and that there can
not be any isolated poles on which magnetic lines of force can
terminate. This equation is
( a ) V . B=O (b)V.D=p

17. The Maxwell's equation that is mathematical expression for

Gauss law is
44 MICROWAVES AND RADAR

18. The Maxwell's equation that is an expression of Ampere's law

for time varying fields is

(C)V . B = O (~)vxH=J+L~
19. One of the Maxwell's equations tells that magnetic voltage
around a closed path equals electric current through the path.
It is

(c)VXE=--
dB
at (d)none of these
20. Maxwell's equation (V x E = - B) when written in differential
form would be

(a)
I I
E.dl=- B.&

Fill in the blanks :

1. Divergence of a vector field (A) is written as
(VA, V .A, V xA)
2. Divergence of a vector field is always a quantity.
(scalar, vector, dimensionless).
3. Magnetic potential is a quantity. (scalar, vector)
4. A static electric field exist in the absence of a mag-
netic field. (can, cannot)
MAXWELL'S EQUATIONS 45

5. A static magnetic field exist in the absence of an

electric field. (can, cannot)
6. With reference to static electric fields, the (i) com-
ponent of (E)is (ii) acmss the charge free interface of
two dielectrics. (normal, tangential, continuous)
7. With reference to static magnetic fields, (i)component
of magnetic flux density is (ii) across the interface.
(tangential, normal, continuous)
8. With reference to static magnetic fields, the (i) com-
ponent of (H) is continuous across a (ii) free interface.
(Charge, current, tangential, normal)
9. The Maxwell's equation V . B = 0 is the outcome of of
a monopole. (existence, non-existence)
-+ 7'
10. The magnitude of cross product of two vectors A and B is
when they are at right angles to each other. (maxi-
mum, minimum, zero)
11. The magnitude of dot product of two vectors is when
the two vectors are in phase. (zero, maximum, infinity)
12. With reference to time varying electric and magnetic fields,
(E)and (H) exist independent of each other (can,
cannot)
+
13. Curl of a vector A is written as . ( V .A, V x A , V'A )
14. The sum of conduction current density and displacement
current density equals . (divergence of H, curl of
H, gradient of H).
15. An example of displacement current is the current apparently
flowing through alan (resistor, capacitor, inductor)

Match the Following

Match the Maxwell's equations given on the extreme left in the
table below with the corresponding word statement from the middle
column and the relevant lawltheorem that the Maxwell's equation
represents from the third column.
46 MICROWAVES AND RADAR

Ih';cxwellEquation I word ~tatern&t Law l Theorem

The e l e d romotive Gauss law

force around a closed
path is equal to the
line derivative of the
magnetic displace-
ment through any
swface bounded by
the path

T h e n e t magnetic Monopole i s non

flux through any existent
closed surfam is zero
The magnetomotive Ampere's law
force around a closed
path is equal to the
sum of conduction
a n d displat ement
current denpities

The total electric Faraday's law

dispiacement
thro&h ~ n surface
y
enclosing a volume is
equei to t h e total
charge enclosed
within the volume
 TRANSMlSSlON MEDIA
- - - - - -

Transmission lines
and Waveguides

oth Transmission lines and Waveguides, the two commonly

B used transmission media at microwave frequencies, have

similar applications, though their operational frequencies are
different with the latter being the preferred choice at higher
microwave frequencies. Transmission lines are used for operational
frequencies equal to or less than about 3 GHz with waveguides replac-
ing transmission lines a t frequencies beyond 3 GHz. The main applica-
tion of transmission lines and waveguides is to carry electromagnetic
energy from one point to another. For instance, a transmission line is
used at the transmitter to interconnect the transmitter output to the
transmission antenna. Another transmission line is used to carry the
electromagnetic energy intercepted by the receiving antenna to the
receiverinput. Other applications of transmissionlines and waveguides
include their use as circuit elements in passive circuits such as filters,
couplers, impedance transformers, resonant circuits, delay lines and so
on.
In this chapter, we shall discuss the operational fundamentals,
typical performance parameters, types and applications of the two
transmission media. The chapter includes a large number of solved
problems to illustrate the concepts discussed in the chapter.
The topics covered in this chapter include :
3.1. Transmission Line 3.5. Impedance Match-
Equivalent Circuit ing
3.2. Sources Of Loss- 3.6. Smith Chart
es in Transmission 3.7. Waveguides
Lines 3.8. Waveguide Modes
3.3. Transmission Line 3.9. Waveguide Para-
Parameters meters
3.4. Types of Trans-
mission Lines
48 MICROWAVES AND RADAR

3.1. Transmission Line Equivalent Circuit

Fig. 3.1. shows the lumped component equivalent network of an
RF transmission line supporting a transverse electromagnetic (TEM)
mode. In the equivalent network shown, R and L are the equivalent

Fig.3.1
series resistance and equivalent seriesinductance respectively, per unit
length of the line. G and C are equivalent shunt conductance and
equivalent shunt capacitance respectively per unit length of the line. In
an ideal lossless transmission line, R = G = 0
3.2. Sources of Losses in Transmission Lines
The three major sources of losses in RF transmission lines are :

Copper losses (also referred to as R losses)

2. Dielectric losses
3. Radiation losses
The CopperLoss or l2I?loss is due to the resistance associatedwith
the conductors constituting the transmission line. This loss appears in
the form of heat. This loss is frequency dependent and increases with
increase in frequency. It may be mentioned here that the resistance
associated with the transmission lines cannot be measured with an
ohmmeter because at radio frequencies, the current tends to flow on the
conductor surface or very near to the conductor surface due to an effect
called skin effect. The effective resistance of the conductor at RF is thus
widely different from its DC resistance.
Dielectric Loss also appears in the form of heat and increases with
increase in frequency. This loss in due to leakage through the dielectric.
Radiation Loss is due to radiation of RF power to fkee space or
nearby circuits. Radiation loss as already pointed out is negligible in
shielded transmission lines like coaxial lines and waveguides.
Although transmission lines are not lossless, yet for all practical
purposes, they can be assumed to be so. This assumption is very much
valid in case of practical transmission lines. The conductors are normal-
ly made up of an easily machinable metal with a coating of silver or any
other highly conducting material. It really doesn't matter what that
TRANSMISSION MEDIA 49
inner material is, as a t R F and microwave frequencies, the current
tends to flow only near the surface. This reduces the copper loss. Also,
in air-filled transmission lines and lines usingloss dielectrics like telfon,
polystyrene, quartz and their mixtures, dielectric loss will also be
negligible, and in shielded lines, radiation loss is insignificant.
3.3. Transmission Line Parameters
Important transmission line parameters include :
1. Characteristic Impedance 2. Propagation Constant
3. Reflection Co-efficient 4. Standing Wave Ratio
5. Input Impedance 6. Return Loss
7. Mismatch Loss
Characteristic Impedance
Characteristic Impedance of a transmission line would be its input
impedance if it was infinitely long. Refer to the transmission line
equivalent circuit of Fig. 3.1. I t can be proved with simple mathematics
that the characteristic impedance of this line is given by :
2, = d(R +joL)/(G +jwC) ...(3.1)
where R = Distributed resistance per unit length
L = Distributed inductance per unit length
G = Distributed shunt conductance per unit length
C = Distributed shunt capacitance per unit length.
In a lossless transmission line, R = 0, G = 0
Therefore, characteristic impedance, Zo =
Characteristic impedance, as is clear from its definition and the
relevant mathematical expression, is characteristic of the line and is
independent of the length of the line. Since all practical transmission
lines are going to be of finite length, the significance of this parameter
arises from the fact that if a finite line is terminated in a load im-
pedance equal to the characteristic impedance of the line, its input im-
pedance in that case will also equal the characteristic impedance.
Propagation Constant
The Propagation Constant (7) is a measure of the attenuation and
the phase shift of the incident waves travelling from the source to the
load end of the transmission line. Propagation constant, for practical
lines, is a complex quantity having a real part known as Attenuation
Constant ( a )and a n imaginary part called Phase Shift constant (P).
The propagation of a wave along a transmission line can be mathe-
matically expressed as
y=&F ...(3.2)
where (2)is the distributed series impedance per unit length and (Y) is
the distributed shunt admittance per unit length. Referring to the
transmission line equivalent circuit of Fig. 3.1.
Z=R+jwL Y=G+,ioC
50 MICROWAVES AND RADAR

Therefore, y = (R +j oL)(G + i w,C,= a +j P

where a = Attenuation on constatt
p = Phase shift constant
Obviously for a lossless transmission h e , R = G = 0
Therefore, y =j a m which implies t h ~at = 0 and f3 = w m.
But
all practical transmission lines do have some loss, however negligible
it might be. As a result, the wave propagating along a trar~smissionline
attenuates as it travels further down the line. If El and E2 are the
voltages at point 1and 2 on a transmission line (Fig, 3.2) then;
E 1 = ea where ( I ) is separation between points 1and 2.
-
E2 1 2

Attenuation in decibels over a length ( I ) is therefore 20 loglo

(E1/E2).The phase constant f3 (Radians per unit length) as expressed
above is directly proportional to frequency. Now the length of the trans-
mission line over which the wave undergoes a phase change of 2n
radians is the wave length (h).
Therefore,
1
which gives phase velocity v,, = h . f = JLC.
Reflection Co-efficient
When a transmission line is terminated in a load impedance which
is not equal to its characteristic impedance, part of the signal energy
sent down the line is reflected back. The ratio of the reflected signal
amplitude to the incident one is defined as the Reflection Coefficient. It
may be expressed as a magnitude only and denoted by (p) or as a com-
plex value having both a magnitude and a phase and denoted by T with
P'I~I.
Mathematically, p= 1r1 =

When the line is short circuited on the load end, i.e. 'ZL= 0
Then p = 1
and when the line is open circuited on the load end, i.e., ZL= m
Then p = 1
TRANSMISSION MEDIA 51

Similarly, when ZL= ij X

Thus there is a total reflection only when the line is either short
circuited on the load end or open circuited on the load end or if the load
is a pure reactance. The three cases are depicted in Fig. 3.3 (a),( b )and
(c) respectively.
Standing Wave Ratio
Whenever a signal travelling along a transmission line comes across a
discontinuity or whenever the line is terminated in a load other than the
characteristic impedance of the line, a part of whole of the incident energy
is reflected back. Under such circuitances, we have two counter propagat-
ing waves in the transmission line. At all those points, where the waves are
in phase, they add producing a signal maximum and at all those points
where they are out of phase, they produce a signal minimum. Thus we have
points of signal maxima and signal minima along the line except for the
case where there is no discontinuity and where the line is terminated in its
characteristic impedance. Since these points of signal maxima and minima
are motionless, standing waves are said to exist.
VSWR, an abbreviation for Voltage Standing Wave Ratio, is the
ratio of Em, to E~,.It is a measure of the mismatch at the discon-
tinuity. Mathematically,
VSWR =
i+lrl
1- l l'-I
VSWR - 1
...(3.7)
'=VSWR+~

VSWR of unity implies a zero reflection co-efficient and thus a per-

fect match. VSWR of infinity implies a unity reflection coefficient and
thus a perfect mismatch.
Input Impedance
The input impedance of a section of a transmission line is given by
equation (3.8)as
(ZL+ Zo tan pl)
Zin = Zo
(2, +j ZL tan pl)
The above expression reveals many interesting properties of open
circuited and short circuited transmission line sections of different
lengths.
52 MICROWAVES AND RADAR

For instance, for a shorted line, ZL= 0.This gives :

Zin =j Z, tan pl
And for a transmission line that is open circuited a t load end,
ZL = infinity. This gives :
Zin = -j Zo cot pl
The input impedance of shorted and open lines as a function of
length, in terms of operating wavelength, are plotted in Fig. 3.4 (a)and
( b )respectively. Following observations are made from these plots :
(a) The input impedance of a shorted line, whose length is odd
integral multiple of (h/4), is infinite.
( b ) The input impedance of a n open line, whose length is odd
integral multiple of (h/4), is zero.
( c ) The input impedance of a shorted line, whose length is in-
tegral multiple of (h/2), is zero which in other words means
that a short a t load end repeats itself a t successive (h/2)
intervals from load end.
(d) The input impedance of an open line, whose length is integral
multiple of (h/2), is infinite. Here the open circuit repeats
itself a t successive (h/2) intervals from load end.
( e ) In general, (ZL) repeats itself a t successive (h/2) intervals
from load end.
( f ) For a line length less than (h/4), the input impedance is
inductive if it is shorted on the load end and capacitive if it
is open on the load end.
Fig. 3.5. shows the impedance offered by different lengths of
shorted and open transmission line sections.
(g)Another interesting observation is that for lines whose length
is odd integral multiple of (3L/4), impedance inversion takes
place from load to the source and such a line can then be used
as a n impedance transformer. I t can be verified that for such
a line,

Since (zo2) is a constant for a given line (Zi,), is inversely propor-

tional to (ZL).

( a )Shorted line ( b )Open line

Fig. 3.4.
TRANSMISSION MEDIA 53

Return Loss
The Return Loss signifies the total round trip loss of the signal and
is defined as the ratio of the incident power to the reflected power a t a
point on the transmission line. It is expressed in decibels.
Return Loss, L,=-2OLogp ...(3.10)
where p = Magnitude of reflection co-efficient
Mismatch Loss
Mismatch Loss is the loss due to reflection from a mismatch. I t is
defined as the ratio of incident power to the difference of incident and
reflected power expressed in decibels.
Mismatch Loss, L, = - 10 Log(1- p2) ...43.I11

I 2
I
I
I
I
I 4t"sse--o
C L
Transm~ssionLlne Sectlon Equivalent c~rcults
Fig. 3.5.

3.4. Types of Transmission Lines

The two commonly used types of transmission lines a t radio fre-
quencies include the open wire lines and t h e coaxial lines. In the
category of open wire lines, the two wire balanced configuration, whose
cross-section is shown in Fig. 3.6 (a),is more common. A coaxial trans-
mission line (Fig. 3.6 ( b ) )comprises a conducting shell, a solid tape or
a braided conductor surrounding an isolated concentric inner conduc-
tor. The inner conductor is either solid or stranded.
54 MICROWAVES AND RADAR

Open wire lines suffer from radiation losses and cross-talk. Radia-
tion losses become prohibitively large a t microwave frequencies.
Coaxial lines, on the other hand, have much better shielding properties
and therefore much lower radiation losses. Coaxial lines are however
unbalanced lines. TEM is the dominant mode. Referring to the two
wire balanced transmission line of Fig. 3.6(a), expressions for comput-
ing important transmission line parameeers are giver. as under:

cha:.acteristic Impedance, Xo =

-r (7)
- 260 loglo - ohms ...(3.12)

Dislributed Inductance,

27.67 E,
Distributed Capacitance, C = In ( W / d ) pF/m

-42ok-
(a) ( b)
Fig. 3.6.
Referring to Fig. 3.6 (b),the relevant expressions for the coaxial
lines are given below :
60
Characteristic Impedance, Zo= In and (b/a) ohm
Er

= 138 loglo (b/a) ohms ...(3.15)

T
Distributed Inductance, L = 200 In (b/a) nfI/m
= 460 loglo (b/a)nH/m ...(3.16)
TRANSMISSION MEDIA 55

55.56 E,
Distributed Capacitance, C= pF/m
L, @/a)
Time Delay, zd = 3.33 c n s / m .,.(3.18)
Cut-off wavelength for higher order models,
h, = rc 6 [a + b] ...(3.19)
3.5. Impedance Matching
Using Transmissioa Lines
Impedance matching is a n important requirement in microwave
circuit design in order to ensure that there is maximum transfer of
power from source to load, that amplitude and phase imbalances are
reduced in power distribution networks and that power loss in feed
lines is minimised. Use of a transmission line to provide an impedance
match involves a transmission line section of characteristic impedance
'Zo'and length '1' depending upon nature of impedances to be matched.
The transmission line section used for matching is connected in either
of the different possible configurations again depending upon matching
requirement. A typical matching problem in practice involves match-
ing a complex impedance, which could be either input or output im-
pedance of a device, to a real impedance. The commonly used
configurations include use of stubs and quarter wave transformers. In
stub matching, again there is a single stub matching technique and a
double stub matching technique. Single stub matching technique uses
either a shunt stub or a series stub. Various techniques outlined here
are briefly described in the following paragraphs.
Single Stub Matching
A stub is basically a shorted or open section of a transmission line
used in conjunction with transmission lines to provide impedance match
and cancel out reflections if any. As the shorted and open transmission
line sections present pure reactances, their introduction doesnot absorb
any power. Fig. 3.7 illustrates the use of a single stub, a shunt stub as it
is connected across the main transmission line, to provide an impedance
match. Here, a transmission line having characteristic impedance of 'Zg'
is shown terminated in a complex load admittance of (gL+jbL). As a h t
step, we locate a point on the transmission line where the normalised ad-

- .yl =l+jbL and y,=-jbL

Fig. 3.7.
56 MICROWAVES AND RADAR

mittance is (1+jbL). It may be mentioned here that ( g +~jbL) is also

the normalised load. In the second step, but a stpb across the transmis-
sion line at that point with the stub designed to offer a susceptance of
(- jbL). Thus the transmission line with a characteristic impedance of
'Zo'gets matched to a complex load. It is a usual practice to use shorted
stubs rather than open ones as it is impossible to get a perfect open. An
open stub if used will always be terminated in free space impedance.
Fig. 3.8 shows the use of series stub. Again, in the first step, we
locate a point on the transmission line where the normalised im-
pedance looking towards the load end is (1+jx). At that point, a stub is
added with the stub offering a normalised reactance of (- jx). As is clear
from Fig. 3.8, the feed line needs to be cut for insertion of series stub.
This technique is therefore not commonly used as it is difficult to fabri-
cate in co-axial and striplines.
L
yd-

z0 -jx
I
r 20

Fig. 3.8.
Double Stub Matching
With the single stub matching of the type discussed above, it is
sometimes impractical to put the stub a t the intended location, more so
in coaxial lines and waveguides. In such cases, double stub matching
technique is preferred. In double stub matching, as shown Fig. 3.9, the
two stubs are put across the main line a t fixed points spaced (3h/8) or
even closer. These stubs have adjustable shorting plungers which can
be adjusted to cancel out most of the reflections.
Shorted
stubs

Transrn~ss~on
l~ne

0
k---'
-or less
8

Fig. 3.9
TRANSMISSION MEDIA 57

Q u a r t e r Wave Transformer
Quarter wave transformer (a h / 4 long line) can be used to match
both a real as well as a complex load impedance to a transmission line. If
the main line characteristic-impedance is 'Zo' and it is to be matched to a
load having a resistance 'RL'the characteristic impedance of the quarter
wave section required for matching would be G. Fig. 3.10 shows the

Fig. 3.10.
interconnection. If the load impedance is complex, say (RL+ j X L ) , it
should first be converted into a real impedance by means of an additional
length 'l' of a line to cancel out the reactive component. If RL' is the real
impedance looking into the input of this additional length towards the
load end, then the characteristic impedances of the quarter wave line sec-
tion is given by :
z , = a
The interconnections are shown in Fig. 3.11. The reactive part of
the load can also be tuned out by using a stub as shown in Fig. 3.12. In
this case, the complex load impedance can be made to present a real
impedance RL' to the quarter wave line section by means of a n (nh/8)
length of a line having a characteristic impedance equal to the mag-
nitude of the load impedance.

Fig. 3.11.
rl

Fig. 3.12.
58 MPCROWAVES AND RADAR

3.6. Smith Chart

Various transmission parameters such a! a theVoltage Standing Wave
Ratio (VSWR),Reflection coefficient (p) and transmission line impedance
f

at a certain point from the load end can be d eteEmined very conveniently
and very quickly with the help of Smith cha -t.In addition to determining
transmission line parameters, smith chart I ?ids extensive application in
working out solutions to impedance matchir ~g problems.
Smith chart is basically an impedance: chart containing two sets of
lines. The first set of lines referred to as Ccmstac~ttResistance lines form
circles (Fig. 3.13), all tangent to each other a t the right hand end of
horizontal diameter. Each circle represents a fixed resistance i.e. all
points on a particular resist,ance circle represent the same resistance.
The resistance values whick~these constant resistance circles represent
are marked on the horizonttal diameter a t the points of intersection of

Fig. 3.13
TRANSMISSION MEDIA 59

these circles and the horizontal line. The outermost resistance circle
cutting the horizontal line on the left extreme represents zero resis-
tance while the circle cutting the horizontal line on the right extreme
represents infinite resistance.
The center of horizontal diameter is labelled as 1. It may be men-
tioned here that all impedances represented on this chart have been nor-
malised with respect to (20).That is, 'l' represents (&). As an example, if
Zo = 50 R, the resistance circle passing through '2' would represent
100R and if (20)was 100 Q, the same circle would represent 200 Q.
There is another set of lines called Constant Reactance lines. These
lines are arcs of circles, all tangent to each other on the right hand
extreme of horizontal diameter and also tangent to this line. The lines
in the upper half represent positive reactances while those in the lower
half represent negative reactances. The reactances have also been nor-
malised with respect to (20).
Though basically an impedance chart, same chart can also be used
a s an Admittance chart alsc. In that case, all points represent admit-
tances. Lines of constant resistmce become lines of constant conduc-
tance and lines of constant reactance become 'lines of constant
susceptance. Again, all admittances would be normalised to (Yo).

Problem 3.1 : When the series resistance R artd the shunt conduc-
tance G of a transmission line are small, but not negligible, determine
expressions for attenuation constant and phase constant.
Solution : y = d(R +j d ) (G +joC) = (a +jp)

and

Problem 3.2 : r h e return loss for a load is 18 dB. Calculate the

Reflection coefficient, 'p' and t h SWR with the load connected.
Solution : LR = 20 log l / p
18 = 20 log l / p
 MICROWAVES AND RADAR

1 18
- = antilog -= 7.94
P 20

S W R = 1.29

Problem 3.3 :A train of pulses with each pulse 30 p wide and

having a n inter-pulse separation of 2 0 p is propagating through a
transmission line. How long should the transmission line be so that ex-
actly three pulses are on the line at a time. Assume propagation speed to
be 3 x lo8 m / s .
Solution :Time duration of the pulse train for having three pulses
on the line at a time = 30 + 20 + 30 + 20 + 30 = 130 ps. This is of course
the minimum time duration required by three pulses.
Therefore, minimum length of cable required
= 3 x lo8 x 130 x
= 39000 m = 39 km

Problem 3.4 :Maximum and minimum values of RMS voltages on

a certain transmission line are 100 V and 25 V respectively. Calculate
the reflection co-efjcicient and characteristic impedance of the transmis-
sion line. Assume that the line is terminated in a load of 300 R.
Solution :Maximum value of RMS voltage = 100 V
Minimum value of RMS voltage = 25 V
100
Therefore, VSWR = -= 4
25
Also, VSWR= 3
20
assuming ZL> Zo

Therefore

:1
Reflection co-efficient, p = ------ -
300 - 75
- 300 + 75
= 0.6

Problem 3.5 :A transmission line with a characteristic impe-

dance of 75R is terminated i n a purely resistive load. While making
measurements, it was observed that the load reflected a power of 100
watts. I f the reflected voltage is 100 V , calculate the load resistance. Also
compute reflection co-effic&nt, the incident power and the power absor-
bed.
Solution : Characteristic impedance of the line. Zo = 75R
TRANSMISSION MEDIA 61

= 100 watts

Therefore,

Reflection coefficient,

Incident power,
I 1-
ZL -20 - 100- 75 1
p= -
ZL+Zo 1 0 0 + 7 5 = ?

Therefore, power absorbed = 700 - 100 = 600 watts.

Problem 3.6 : A lOOR transmission line is to be matched to a

25Q antenna with apiece of transmission line. If the operating frequen-
cy is 100 MHz, calculate the length of the line required and also the
characteristic impedance of the line from which it is to cut.
h
Solution :Length of line required = -
4
where

Therefore, length of line = 3/4 = 75 cm

Characteristic impedance of the line from which the quarter wave
length is to be cut is given by :
-= OR

Problem 3.7 : A 75 R transmission line that is half wave long is

terminated i n a load resistance of 300 R. Determine its input i m -
pedance. If the same line is then operated a s a frequency half o f the
original operating frequency, then what would be its effect on the input
impedance ?
Solution : In the first case when the line is (h/2) long, the input
impedance is same as the load resistance i.e. 300 Q. Input impedance
is, in general, given by

For
Therefore, tan Dl = 0 which gives
Zi= ZL From (1)
62 MICROWAVES AND RADAR

In the second case when the operating frequency is halved, the

wavelength is doubled which means that the same line is now (114)
long.
For
Therefore, tan Pl= =.
From ( 2 ) ,we get

Problem 3.8 : Refer to the transmission line section shown in Fig.

8
3.14. If the propagation velocity is 2 x 10 m / s , write expressions for
V i n ) and (VL)

100 cos wt

Fig. 5.14

2 x lo8
Solution :f = 100 MHz, A = -- =2m
10.3 x lo6
h
Therefore, the given line is - long which implies that
2

loo 300
vin = (300+ 300) cos 2x x 10%
= 50 cos 2 ~ x 10%
Since the line is ( h / 2 ) long, the signal undergoes a phase delay of
pl = (2n/h)x h / 2 = n radians as it travels to the load end.
Therefore, V L = 50 cos (2x x lo8 t - K ) .
3.7. Wave Guide
A waveguide does the same job a t microwaves which the transmis-
sion lines usually do a t relatively lower frequencies. At microwaves, it
is more convenient to talk in terms of electric and magnetic fields
propagating in the transmission medium rather than voltages and cur-
rents which we are familiar with in case of transmission lines. At rela-
tively lower frequencies extending upto say 100 MHz or so, AC circuit
TRANSMISSION MEDU 63

t h e o q is very,weU8devel oped and almost all electronic functions can be

implemented with av, ailable lumped components like Resistors,
Capacitors, Indudors e tc. interconnected with wires to form a circuit.
This, approach d' A.C. . circuit theory, which is nothing but a n ap-
proximation to the fielc i theory explained by Maxwell's equations and
the Wave Equation, ho wever breaks down as we operate a t higher fre-
quencies exceeding 1 GHz or so. At those frequencies, it is more
relevant to talk in terr ns of electromagnetic fields. The reason for this
is very simple. At mi( xowave frequencies, where the cororesponding
wavelengths a r e typic ally few tens of centimeters or lower, the size of
lumped circuik elem1 ents and interconnecting wires becomes com-
parable to the wavele ngth. Due to this, the electromagnetic energy in-
stead of remaining co nfined to the circuit gets radiated.
Waveguide is th e transmission medium of choice a t higher fre-
quencies.
A waveguide is r lothing but a conducting tube through which ener-
gy is transmitted in the form of electromagnetic waves. The waveguide
can be considered t o be a boundary which confines the waves to the
space enclosed by b oundary walls. An ideal waveguide would perform
this task without 1oss of energy or any distortion of the propagating
wave. Actual waveguides, however, only approximate to this ideal con-
dition. The waveguide can assume any shape theoretically but the
analysis of irregularly shaped guides becomes very difficult. Two
popular types are the rectangular waveguides and circular waveguides
and again out of the two, former is more extensively used. Fig. 3.15 (a)
and ( b ) show t h e outlines of rectangular and circular waveguides
respectively.
R rectangular waveguide is characterised by its wide dimension ( a )
and narrow dimension (b) whereas a circular waveguide is charac-
terised by the internal diameter ( d ) .

3.8. Waveguide Modes

There will in general be infinite number of possible electric and mag-
netic field configurations inside the waveguide if there was no upper limit
for the frequency of the signal to be transmitted. Each of these field con-
figurations is called a Mode. There are two types of modes; i.e. TM modes
(TransverseMagnetic)and TE modes (Transverse Electric). In TM modes,
magnetic lines are entirely transverse to the direction of propagation of
e.m wave. The electric field has a component in that direction. In TE mode
(Transverse Electric), the electric field lines are entirely transverse to the
direction of propagatian whereas magnetic field has a component along
the direction of propagation. Various propagation modes, both TM and
TE, are designated hy two subscripts. The first subscript indicates the
number of half wave variations of the electric field in the wide dimension
of the waveguide whereas the second subscript indicates the number of
half-wave variations along the narrow dimension of the waveguide. For
instance, in TElo, mode, which is the simplest mode, there is only one
64 MICROWAVES AND RADAR

half-wave variation of electric field along the wide dimension and there
is no electric field variation along the narrow dimension. Refer to Fig.
3.16. It may be mentioned that this subscript notation is only for rec-
tangular waveguides. In circular waveguides, the subscripts are there
but they do not carry the same meaning as they do in case of rectan-
gular waveguides. Waveguide modes in rectangular and circular wave
guides are discussed further in the latter part of the chapter.

Fig. 3.15

Fig. 3.16
Dominant Modes
Dominant mode propagating in a waveguide is one which has the
highest cut-off wavelength for a waveguide of given dimensions. The
cut-off wavelength of a waveguide is the highest signal wavelength
that can propagate in a given waveguide. I t is discussed in detail in the
next section. It will be seen that TElo mode is the dominant mode in
rectangular waveguides. Now if we choose the guide dimensions in
such a way t h a t t h e signal wavelength is less than the cut-off
wavelength for TElo mode and greater than the cut-off wavelength a t
all other modes, which is easily achievable, we can ensure that only
TElo mode propagates. That is why, TElo mode is called the dominant
mode. Even if a higher mode gets excited due to a discontinuity in the
waveguide, it would soon die out as the guide would not support that
mode. I t will also be seen, that TEll mode has the highest cut-off
wavelength in a circular waveguide and we can always choose a
diameter so that only TEll mode propagates. This should then be the
dominant mode in circular guides. However due to unsymmetrical na-
ture of this mode [see Fig. 3.17 (a)]and due to symmetrical nature of a
circular guide, this mode is not the most popular a s a bend or a discon-
tinuity in the guide might twist the mode leading to propagation with
wrong polarisation. TMol and TEol modes on the other hand, are sym-
metrical [Refer to Fig. 3.17 (b)l.TMol mode is used where symmetry is
important whereas TEol is used for long distance waveguide runs us it
has the least attenuation of all the commonly used modes in circular
TRANSMISSION MEDIA 65

waveguides. Also its attenuation decreases a frequency increases and

is thus useful a t higher microwave frequencies.

(b)
Fig. 3.17
TEM Mode
In a TEM wave, both electric as well as magnetic field are entirely
transverse. It is significant to note that a TEM mode can not propagate
in a hollow waveguide. Let us assume that the magnetic field is entire-
ly transverse, this means that the magnetic field lines must entirely be
in the transverse plane. Also in a non-magnetic material the lines of
magnetic field must form closed loops. Thus if a TEM wave exists in a
waveguide, the first conditicn to be met is that lines of magnetic field
will be closed loops in a plane perpendicular to the propagation axis.
According to Maxwell's first equation, magnetomotive force around
each of these closed loops must equal the axial current, conduction cur-
rent or displacement current, through the loop. Now there cannot be
any conduction current, the wave guide being hollow, and there can be
an axial displacement current only if there is an axial component of
electric field which is again not there in TEM wave. Hence our assump-
tion that a TEM wave exists in a hollow waveguide is fundamentally
wrong which implies that a TEM wave cannot propagate in a hollow
waveguide.
However, TEM mode is the principal mode in coaxial lines. The
electric field lines extend from one conductor to the
other as shown in Fig. 3.18 and are wholly transverse
with no component in the direction of propagation.
The magnetic lines are closed loops and have no com-
ponent in the direction of propagation. It has no cut-off
wavelength and can propagate at all frequencies right
from dc to microwaves. You would notice that in this
case there is no violation of any electromagnetic laws Fig. 3.18.
for TEM wave to exist as in this case the axial current
can flow through the central conductor.
3.9. Waveguide Parameters
Important waveguide parameters include : .
1. Cut-off wavelength 2. Guide wavelength
3. Group and Phase velocities 4. Characteristic wave impedance
66 MICROWAVES AND RADAR

Cut-offWavelength
As already outlined, there are a nur~lberaf possible electric field
and magnetic field co.~figurations (called Modes) that c m exist in a
waveguide. The modes that can exist and sustain in a waveguide are a
function ofwaveguide dimensions and the frequency of the propagating
signal. Each mode has a cut-off wavelength ;.e. for a particular mode to
sustain, the wavelength corresponding to the signal frequency must be
less than the cut-off wavelength for that mode. The cut-off wavelength
for rectangular guides for both TE,, and TM,, is given by

' z
= \l(rn/a)2+ (n/b)2
where a =Wide dimension of waveguide
b = Narrow dimension of waveguide.
Table 3.1 enlists the cut-off wavelengths for various common mode
in rectangular waveguides.
Table 3.1 Table 3.2

1 Mode 1 cut-0ff
wavelength I I Made I cut-off
wavelength

TE,, or TMll 2ab T"o~ 1.306 d

G=z7

The cut-off wavelengh of a circular waveguide with an internal

diameter of (dl is given by :

[a
h (Circular Waveguide) = -
, ,
where K,. = Solution of a Bessel function equation. The values of (K,) for
TEol, T E l l , TE21, TEo2, TE12 and TE22 modes are 3.83, 1.84, 3.05,
7.02, 5.33 and 6.71 respectively. The values of (K,) for TMol, TMl1,
TM21,TMo2, T M I 2and TM22 modes are 2.4, 3.83, 5.14, 5.52, 7.02 and
8.42 respectively.
Table 3.2 gives cut-off wavelengths for various common modes in
circular waveguides. (d) is the internal diameter of the guide.
Guide Wavelength
Guide wavelength, wavelength of the travelling wave propagating in-
side the waveguide, is always different from the free space wavelength
TRANSMISSION MEDIA 67

(A). The guide wavelength (Ag),the cut-off wavelength (A,) and the free
space wavelength (A) are interrelated by :

This expression is valid for any waveguide mode and cross-section provided
the value of (h) substituted corresponds to that mode and cross-section.

From the above expression, if h < < h, ,

Physical interpretation of the concept of Guide Wavelength (G)is

discussed further in the next section.
Group Velocity and Phase Velocity
The velocity of propagation in a waveguide is the product of guide
wavelength (Ag) and frequency (f) and since Ag > A, it appears as if (up)
is greater than speed of light. This appears to contradict the law that
no signal can be transmitted faster than the speed of light. In
waveguides also, it is found that intelligence or modulation does not
travel at a velocity (up). (up) is termed as the velocity of phase or Phase
Velocity. When a modulated carrier travels through a waveguide, the
modulation envelope travels with a velocity much less than that of car-
rier and even less than the speed of light. The velocity of modulation
envelope is called Group Velocity (vg) As ug is less than up, the modula-
tion keeps slipping backwards with respect to carrier as the modulated
signal travels in a waveguide.
In an air filled or hollow waveguide, the phase and group velocities
are related to speed of light by

and

so that c 2 = up vg . For a waveguide filled with dielectric having dielectric

constant (e,.),

c2 = u p ug ...(3.21)
Concept of Group and Phase Velocities is discussed further in the
next section.
88 MICROWAVES AND RADAR

Characteristic Wave Impedance

The generalised expression for the characteristics impedance (20)
of waveguide for TE modes is :

\ /
The generalised expression for the characteristic impedance (20) of
waveguide for TM modes is :

\
where p = Relative permeability of wa<eguide material
= unity for nonmagnetic materials
E =Relative permittivity of the dielectric used
= unity for air filled or hollow guides
For rectangular waveguides, a = wide dimer sion, b = narrow
dimension. For circular waveguides, a = b.
(i) For hollow or air filled waveguides, E = 1,p = 1
Therefore, Zo = 377 For TE modes
b h
Zo=377-- For TM modes
ak
(ii) For waveguide filled with a dielectric material of dielectric con-
stant (&),
For TE modes

For TM modes

(iii) For hollow or air filled circular waveguides,

Zo =(377 *) For TE modes

For TM modes
(iv) For circular waveguides filled with dielectric material of
dielectric constant ( E ) ,

For TE modes

For TM modes
TRANSMISSION MEDIA 69

The characteristic wave impedance for TE,, modes is related to

the free space characteristic impedance by :

The characteristic wave impedance for TM,, modes is related to

the free space characteristic impedance by :
z0= zJ[J
- ...(3.25)
where Z = Characteristic impedance of free space = 377 Q
h, = Cut-off wavelength

Problem 3.9 :Show that for a TElo mode, a frequency of 8 GHz

will pass through a wave-guide of dimensions a = 1.5 cm, b = 1 cm i f a
dielectric with E, = 4 is inserted inside the guide.
Solution : Since a = 1.5 cm, cut-off wave-length for TElo mode
hc=2a=3cm.
3 x 1o1O
Wavelength corresponding to 8 GHz = ------= 3.75 cm
8 x lo9
Since the wavelength to be propagated is higher than the cut-off
wavelength of 3 cm, the signal will not propagate.
When the waveguide is filled with dielectric,
hair 3.75
Then, hdiekctric
= = = 1.875Cm
E
which is less than 3 cm and hence 8 GHz frequency will pass through
the guide.
-

Problem 3.10 : A rectangular waveguide has wide and narrow

dimensions of 4 and 2 cms respectively. What is the minimum frequency
which can be transmitted if a dielectric of E, = 4 be inserted inside the

Solution : a = 4 cm
b=2cm

Minimum frequency which can be passed without dielectric is

3.75 GHz.
70 MICROWAWS AND RADAR

But if we insert a dielectric

M i n i w ~ mfrequency that can be passed with dielectric in the

waveguide is therefore 7.5 GHz.

Problem 3.11 : Show that a 1 GHz signal cannot propagate in the

TElo w ~ d in
e a rectangular waveguide havingj a wall separation of 5 cm.
Solution :The cut-off wavelength (b) for TE,, mode is given by
2
ho =
d(m/a12+ (n/b12
where a =wall separation
2a
ho = -
m
..............n = 0 for TElo mode

3 x 1o1O
Free space wavelength = - = 30 cm
lo9
As the free space wavelength is greater than the cut-off wavelength
for TElo mode, this mode c a n ~ opropagate
t in the given waveguide.

Problem 3.12 :Refer to Fig. 3.19. Determine the longest cut-off

wavelength for this waveguide.
Solution :

L 3 5 m r n d

Fig. 3.19.
For the given waveguide, a = 30 mm, b = 20 mm
The longest cut-off wavelength is for TElo mode also called the
dominant mode and is given by
TRANSMISSION MEDIA 71

3.10. Plane Wave Propagation in Waveguides

To illustrate further some of the concepts and the parameters out-
lined earlier and to get an insight of what happens inside a waveguide
as a plane wave propagates through it, we shall take the case of a plane
wave propagating in a rectangular waveguide. Detailed mathematical
treatment is avoided here and the emphasis would be on under-
standing the physical concepts behind waveguide's ability to guide e.m.
waves. The results obtained and the conclusions drawn in this section
will be further validated by a comprehensive mathematical analysis
based on solving the Maxwell's equations and the wave equation.
Fig. 3.20 shows the cross-section of a rectangular waveguide with
(a)and (b) as its wide and narrow dimensions. As the waveguide is
made up of a good conducting material such as coppel;, a plane wave
solution, in which E and H a r e uniform, is not possible across the guide
if the boundary conditions are to be satisfied. According to the bound-
ary conditions, a t the surface of a perfect conductor, the tangential
component of electric field and normal component of magnetic field
should be zero. Thus, we try satisfying the boundary conditions on the
top wall, we tend to violate them on the side walls. The vice versa is
also true.

Fig. 3.20.
Such a situation can also be considered as a corollary of the fact
that a TEM (Transverse Electric Magnetic) wave cali not propagate
inside a hollow single conductor waveguide. This has been discussed
and verified in the earlier part of the chapter. However, it can be
proved that a plane wave can propagate through the waveguide by
bouncing back and forth between the side walls of the guide. Refer to
Fig. 3.21. Let a plane wave be obliquely incident on one of the side walls
of the waveguide and assume E-vector to be normal to the plane of
incidence as shown in the diagram. (A) here is a typical point on the
wavefront. Consider planes perpendicular to the plane carrying inci-
72 MICROWAVES AND RADAR

Perfect
Conductor

Fig. 3.21.
dent and reflected rays at intervals of (11/4)from (A).The incident ray
undergoes a phase change of 180"or ( x ) radians after reflection accord-
ing to the requirements of the boundary conditions at the surface of
conductor. Therefore, planes lying on either side of point (A) and
separated by integer number of wavelengths along OAO will be ( x )
radians out of phase. The points where such wavefionts interact, the
result is a zero total field. Fig. 3.22 illustrates the point further. BB' is
one such plane formed by intersecting wavefronts that are one full
wavelength apart. CC' is another formed by intersecting wavefronts
that are (211)apart. What is interesting and of practical significance

'A,
Fig.3.22.
TRANSMISSION MEDIA 73

here is that since the field is zero on these planes (BE,CC'),we can add
another conducting surface in this plane without violating the bound-
ary conditions and without altering the field pattern in the region be-
tween the conductors. This further implies that the reflected wave from
the newly added conducting plane will exactly replace the incident
wave. That is, once launched, a wave can reflect successively between
parallel conducting walls without violating boundary conditions on
either side. The tlow of electromagnetic energy between the two con-
ducting planes is depicted in Fig. 3.23. If these planes constitute the
side walls of a rectangular waveguide, the electromagnetic energy
bounces back and forth fr$m the side walls while propagating through
the waveguide. The electric field in the region between the two side
walls of the waveguides, will be identical to the superposition of two
plane waves. And since electric field is normal to the plane of incidence
(Fig. 3.211, top and bottom plates can also be added to complete the
waveguide structure without affecting the field pattern and the flow of
electromagnetic energy through the waveguide.
/Side Wall .

Side wall/
Fig. 3.23.
With this background, we shall now discuss the phenomenon of
cut-offwavelength in a waveguide. We know that a waveguide behaves
like a high pass fitter and frequencies lower than the waveguide cut-off
frequency or wavelengths larger than +he cut-off wavelength can not

Fig. 3.24.
74 MICROWAVES AND RADAR

propagate. We also know that there are various modes of propagation

of electromagnetic waves in waveguides and each one of the modes is
associated with its own cut-off frequency or wavelength. These have
been outlined earlier and shall be further discussed in the latter part
of the chapter. For the time being, we shall focus on the phenomenon
responsible for existence of a cut-off wavelength. Refer to Fig. 3.24. It
depicts two of the many wavefronts of Fig. 3.22. These two wavefronts
are (A)apart and are located on either side of (A).
From the figure, from triangles APP' and A ' P '

AIy= cos ( 2 a - 90)

h
AP'=
cos (2a - 90)
a
--
Also, AP' - cos (90 - a)

a
or M=
cos (90 - a)
- a
which gives,
cos (2a - 90) - cos (90 - a)
h - a
sin 2a sin a
h - sin 2a - 2 sin a cos a = cos a
a sin a sin a
cos a = I-$[
\ /
Now what does this simple deduction tell ? Since (cos a) can not be
greater than unity, therefore, (A)can a t the most be equal to (2a).Thus
any (A) greater than (2a)is not permitted. For a waveguide of a given
with (a),(2a) becomes the cut-off wavelength (A,). The corresponding
frequency fc = 5 then is the cut-off frequency. Thus s t q t i n g with the
kc
basic conditions for propagation inside a waveguide, we have been able
to establish the existence of a certain cut-off wavelength. For
wavelength equal to the cut-off wavelength, angle (a) becomes zero
which means that the wave bounces back and forth between the side
walls of the guide a t the same point. It may also be mentioned here that
h, = 2a is the highest possible cut-off wavelength for the rectangular
waveguide and this highest cut-off wavelength occurs for the dominant
mode.
The field pattern within the guide can be determined by adding
incident and reflected waves. Figs. 3.25(a)and (b) depict two ways of
showing the total electric field. While in Fig. 3.25(a), length of arrows
represents field strength, it is the density of arrows in case of Fig.
3.25(b).
TRANSMISSION MEDIA 75

Fig. 3.25.

If we look along the waveguide, the situation is like what is shown

in Fig. 3.26.If the field pattern is the one of Fig. 3.25(a) at time t = 0,
then the field patterns at t = 7/4 (Quarter period) and t = 3T/4 (three
quarters of one time period) will be zero. At t = 7/2 and t = T, the field
patterns would be as shown in Fig. 3.26.Fig. 3.26also brings to the fore
the concept of Guide Wavelengtn (Ag). The geometry shown in Fig. 3.24
can be used to determine the relationship betw: en free space wavelength
(h) and guide wavelength (Ag).

Fig.3.26.
From Fig. 3.24, h = kg sin a
76 MICROWAVES AND RADAR

Having computed the wavelength inside the waveguide, the phase

velocity (up)of the field pattern can be computed from,
U
J?
kg
= which gives up = k]e
Though this gives an impression that velocity of propagation inside
the waveguide is greater than free space velocity, it is not so in reality.
The electromagnetic energy flows inside the guide as a wave travelling
a t an angle of (90-a) degrees to the direction of propagation. The
resultant velocity of propagation is termed as the group velocity (vg)
and is given by,

ug = c cos (90- a)= c sin a = - c

Expectedly, (vg) is less than (c). The product'ofphase and group

(4
velocities yields,

// .
,---
,' ,.--.
0
\
,' Q /*---. 0
0
0
I
- -
t0 \

, '\
\ \
; Q @
/"
/'
\
\, 0 0
Electric
o.o!tt \ \\ .--'
? ! !, e z ~ ! ! t ) + ! a 8
/
/
\
\
\ - 4
/ /
@ \,
x.
'.*---*
/
,/
8 ,
\.'
\
.---I
0
,
/
0

\Mapanetic Field
Fig. 3,27(a).
Till now, we have been talking about the electric field. The magnetic
field can also be sketched in a similar
manner keeping in view the fact
that direction of (E) and (El) is such
t h a t t h e Poynting vector (E x H)
gives the direction of energy flow.
The electric and magnetic fields in-
side the guide as viewed fi-om the top
along the guide length and as viewed
from the end are respectively shown
in Figs. 3.27(a) and (b). Fig. 3.27(b).
TRANSMISSION MEDIA 77

3.1 1. Transverse Electric Modes in Rectangular

Waveguides - Mathematical Analysis
The electromagnetic field configuration within the waveguide can
be obtained by solving the Wave Equation and the Maxwell's Equa-
tions subject to appropriate boundary conditions. We shall assume that
the walls of the waveguide are made up of a material that is a perfect
conductor i.e., it has infinite conductivity. Though the waveguides are
usually made up of copper or brass and these two materials are not
perfect conductors yet our assumption is very safe. I t is very easy to
prove that for good conductors such as copper or brass, their finite con-
ductivity has negligible effect on field configuration. Finite conduc-
tivity leads to surface currents flowing in these conducting planes
constituting waveguide walls. The surface currents can be used to com-
pute losses and subsequently attenuation of the wave. The waveguide
losses have been discussed in detail later.
Also, assumption of infinite conductivity leads us to simple bound-
ary conditions stated as under :
1. Tangential component of Electric field is zero a t the surface of
the conductor.
2. Normal component of Magnetic field is zero a t the surface of the
conductor.
For the purpose of mathematical analysis, Maxwell's equations
and the wave equation are expressed in rectangular co-ordinates for
rectangular waveguides. Refer to Fig. 3.28. Z-axis is the direction of
propagation. In a TE (Transverse Electric) wave, the electric field is
wholly transverse to the direction of propagation. That is, there is no
component of electric field in the direction of wave propagation which
implies that Ez= 0. However, magnetic field does have a component
(Hz) in the direction of wave propagation.

Direction Of
Wave Propagation

F
'
z
Fig. 3.28.
The wave equation for the 2-component of electric field can be writ-
ten as,
V ~ H=, - O ~ ~ E H ,
78 MICROWAVES AND RADAR

In rectangular co-ordinates, the equation becomes

a - - y. The above equation reduces to

Let us define an operator, -
az -

where h2 = (y + W ~ P E )
h2 is a real number whose value d e p a d s upon waveguide dimensions
and the order of mode being consider3d.
Also, Y = J o W ) (0 +JW)
where the terms have their usual mcaning. As the space between the
conducting planes is non-conducting, o = 0 which gives
y=6m=dG&F=jco~
or p =-W ~ P E
Equation (3.26)is a partial d~fferentialequation with a solution of
the form
H,=X. Y
where X = pure function of x only
Y = pure function of y only
Substituting Hz=XY in equation (3.26),we get

Assume 1 d2x
x z = - ~ 2

and
TRANSMISSION MEDIA 79

) (- B ~are
where ( - A ~and ) constants. Equation (3.27) reduces to
h2 = A ~ + B ~ ...(3.28)
Now solving for (X) and (Y) by separation of variable method, we
get
X = C 1 cos Bx+C2sin Bx
and Y = C3 cos Ay + C4 sinAy
where C1, C2, C3 and C4 are constants.
Substituting for (X) and (Y) in Hz = XY,
Hz = (C1 cos Bx + C2 sin Bx) (C3 cos Ay + C4 sin Ay) ...(3.29)
Constants C1, C2, C3 and C4 can be determined by applying ap-
propriate boundary conditions to equation (3.29).
The boundary conditions are as follows :
1. The component of electric field (E,) in the direction of
propagation (i.e. 2-direction) is zero. That is E, = 0
2. The component of electric field (E,) is zero in y = 0 and y = b
planes. Refer to Fig. 3.28, y = 0 and y = b planes correspond
to bottom and top walls of the waveguide.
3. The component of electric field (E,) is zero in x = 0 and x = a
planes. Refer to Fig. 3.28. x = 0 and x = a planes correspond
to left and right side walls of the waveguide.
It may be mentioned here that Condition-1 is the outcome of the
wave being transverse electric in nature. Conditions 2 and 3 conform
to the boundary condition that tangential component of electric field is
zero. So, E, is non-zero as for as the side walls are concerned and Ey is
non-zero in top and bottom walls.
As seen above, the boundary conditions involve (Ex)and (E,).
Before proceeding further, therefore, we shall first determine expres-
sions for (Ex) and (Ey). We shall :hen apply the boundary conditions
outlined above to determine constants C1, C2, C3 and C4 which in turn
will lead us to a n expression for Hz.
Determination of Expressions for E, and Ey, H, and Hy
Maxwell's curl equations involving (H)and (E)are written as :
VXH=(G+JWE)E
and VxE=-JWW
If the region between the conducting planes is non-conductive,
these equations become
VxH=jorE ..i3.30)
and vxE=-j~@ ...(3.31)
80 MICROWAVES AND RADAR

The relevant wave equations are

V ~ =E YE ...(3.32)
and V ~ =H YH ...(3.33)
where y = Propagation constant = d(o +j w ) (imp)
Expanding equation (3.31),
If f $ 1

Replacing ( a / & ) by (- y) and expan+qg the determinant on L.H.S.

A,
and then comparing the co-efficients of r , j and we get

and
Expanding equation (3.30)and going through similar operations
would lead us to following three equations :
2%
-+ yHy =j o S x ...(3.37)
?Y

aHY aHx =jwEEZ

--
and -7
ax ay
Substituting the value of (Hy)from Eqn. (3.35)in Eqn. (3.37)and
simplifying, we get the following expression for (Ex),

Substituting the value of (H,) from Eqn. (3.34)in Eqn. (3.38),we

get the following expression for (EY),

Expressions for (Hz)and (Hy)can similarly be found out to be as,

and
TRANSMISSION MEDIA 81

Expressions for (H,) and (Hy) will be used when we discuss TM

(Transverse Magnetic) waves.
Coming back to our objective of solving for C1, C2,
Cg and C4,

--
* J (Cl cos Bx + C2 sin Bx) (C3cos Ay + C4 sin Ay)
h2 ay
- a(C1cos Bx + C2 sin Bx) (- AC3 sin Ay + AC4 cos Ay)
h2
Substituting the boundary condition, Ex = 0 for y = 0, we get
(C1cos Bx + C2 sin Bx) (0 +AC4)= 0
Since (C1cos Bx + C2 sin Bx) # 0 and A # 0
This gives C4 = 0. Putting C4 = 0 in Eqn. (3.291, we get
Hz= (C1cos Bx + C2 sin Bx) (C2cos Ay) ...(3.44)
According to another boundary condition, E,, = 0 a t x = 0

- A [( c1cos BX + c2sin B X )(c3cos ~ y ) l

* J
h2 ax
I(- BCl sin Bx + B C cos Bx) (C3cos Ay)]
h2
Substituting, Ey = 0 a t x = 0, we get

(BC2C3 cos Ay) = 0

h
Since cos Ay # 0, B # 0.This gives C2 = 0.
Substituting the value of (C2)in Eqn. (3.44),we get
Hz = C1C3cos BX cos Ay ...(3.45)
From the expression ofE,,

E x = -- Y - j w aHz = -JV. aH,

h2 ax h2 ?Y h2 ay
- (C1C3cos Bx cos Ay)
h2 ay
- * h2
(- clc3A cos BX sin ~ y )

= J*
h2
(ACIC3cos Bx sin Ay)
82 MICROWAVES AND RADAR

Substituting the boundary condition, Ex = 0 a t y = b, we get

@ AC,C3 cos Bx sin Ab = 0

h2
Now cos Bx + 0, C1,C3 + 0,Therefore sin Ab = 0
or Ab = nn where n is a n integer.
or A = (nsc/b)
From the expression of Ey,

(- BC1C3 sin Bx cos Ay)

h2
- 2SE
!(BC1C3sin Bx cos Ay)
h2
Substituting the boundary condition, Ey = 0 a t x = a ,

-lop(BC1C3sin Ba cos Ay) = 0

--

h2
Since cos Ay # 0,C1,C3 + 0,B # 0
Therefore, sin Ba = 0
or Ba = msc
or B = ( m x / a )where ( m )is a n integer
Thus we have determined
A = ( n n / b ) and B = ( m n / a ) ...(3.47)
Substituting for ( A )and ( B ) in Eqn. 3.45 to get complete solution
for Hz as
Hz = C1C3Cos Bx COS Ay
= C1C3cos (mrc/a)x cos ( n x / b )y

Let C1C3= C where (C)is another constant,

Hz = C cos ( m x / a )x cos ( n n / b )y ...(3.48)
Since the propagation is in 2-direction, various field components
will follow exponential relation (e'-Yz)in this direction. And if the time
variation factor is combined with 2-variation factor, it is seen that
(e-P . dot)= euwt- represents a wave propagating in the 2-direction.
Thus, expression for ( H z )as given by Eqn. (3.48)can be written as,
H,= C cos ( m x / a ) xcos ( n x / b ) yeuwt ...(3.49)
TRANSMISSION MEDIA 83

The expressions for other field components are derived as under.

From Eqn. (3.40),
-y aE* - jwp 2%
X - h 2 ax h2 ?Y
-
-- a
h2 ay
[C cos (7) (7)
y eoa
x cos - y)]

=-JT '

h2
[ccos[T]
-x .[<)sin(~)ye~d-~)]

~ ~ C ( ~ ) [ ~ ~ ~ ( ~ ) x s i n ...
E = h2 ( (3.50)
~ ] ~ ] e ~ ~ ~ ~

From Eqn. (3.41)

84 MICROWAVES AND RADAR

h 2 [ nx
=-L b ) [ [e] . .

Ccos - xcos ( : ' ) y ] e o o . - p )

. .

To sum up expressions for Ex, Eyl H,, Hy and Hzfor a transverse

electric (TE)wave propagating in a rectangular waveguide in Z-direc-
tion are given by equations 3.50, 3.51,3.52, 3.53 and 3.49 respectively
and are reproduced below.

H,=-c $ ( - ) (T) (T)

- ]
sin - xcos y eG".-F)

The process of determining mathematical exmessions for various

field components Ex,Ey, H,, 4 and Hz (E, being zero for a TE wave
propagating in the z-direction) as outlined in the previous is apparently
cumbersome. But it is certainly not complex. If we remember the curl
Maxwell's equations and the wave equations along with boundary con-
ditions, the rest of it follows automatically. So, all that one needs to
remember is the following :
1. V x H = (0 +jw) E, (T = 0 for non-conducting region
2. V x E = - j o @
YH
3. v 2 ~ =
YE
4 . V ~ =E
5 . Tangential component of ( E ) and Normal component of ( H ) are
zero at the surface of a conductor.
3.12. Various TE Modes in Rectangular
Waveguides
Different TE modes can be determined by substituting different values
for m and n in the generalised form TE,,. (m) and ( n )are 0 , 1 , 2 , 3 ....
For m = 0,n = 0 , we get TEoo mode. It can be seen that all field com-
ponents (Ex,E,,, H,, Hy)are zero for TEoo mode. Therefore, this mode does
not exist.
TRANSMISSION MEDIA 85

For m = 0, n = 1,we get TEol mode. For this mode, Ey = 0, H, = 0

Ex and Hy exist and are given by :
Ex = (jwp)/h2 C (n/b) sin (n/b)y eGa -p)
Y C (n/b) cos (n/b)y eti*-p)
Hy = - -
h2
The cut-off wavelength for this mode can be determined from
n n

For m = 1,n = 0,we get TElo mode. For this mode, Ex = 0 , Hy = 0,

Ey and H, exist and are given by :

E
- j u p C JC sin
~ ( a )= (E) xetiot-p)
~

The cut-off wavelength for TElo mode can be determined from :

2 - 2a
h, =
d(m/a12 + (n/b)2
For m = 1,n = 1, we get TEll mode and for this mode Ex, Ey, H,
and Hy exist and so is the case for all other higher order modes. The
cut-off wavelength for TEll mode is given by :

It can be seen that TElo mode has the highest cut-off wavelength
(= 2a) in case of rectangular waveguides. That is why, it is also known
as the Dominant Mode. The field pattern for TElo mode was shown
earlier in Fig. 3.16.
3.13. Transverse Magnetic (TM) Waves in
Rectangular Waveguides
- Mathematical Analysis
The procedure for determining expressions for various field com-
ponents (E,, Ey, E,, H,, Hy)isjust the same as outlined in the previous
pages in the case of TE waves. Of course, in case of TM waves, it is Hz
(and not E,) that is zero,
That is Hz = 0 and Ez # 0
aE,
The wave equation is -+ 7+ h 2 ~ =, 0
ax2 ?Y
This equation can then be worked on along with Maxwell's curl
equations to get an expression similar to Eqn. 3.29 obtained in case of
TE waves. The equation (for TM waves) would be as given under,
86 MICROWAVES AND RADAR

E, = (C1cos Bx t C2 sin Bx) (C3cos Ay + C4 sin Ay) ...(3.55)

Solutions for various field components can then be determined by
applying appropriate boundary conditions to Eqn. 3.55 and then solv-

-
ing fur C1,C2, C3 and C4. The boundary conditions to be applied here
incladc E, = 0 for y = O and y b (Bottom and Top walls) and E, = 0 for
x =- 0 and x = a (left and right side walk). The final expression for (E,)
can be deiermined to be,
E, .
= C sin (mn/a)x sin (nn/b)y etiOt-P) ...(3.56)
I'he expressions for the field components are a s under,
Ex = -*I C(mn/a)cos ( r n d a )x sin (n.rc/b)y etiW-P) ...(3.57)
h2
Ey= 2
h
C(nn/b)sin ( r n d b )x cos (nn/b)y etiWt ...(3.58)

i-I,= J-7 C(nn/h) sin ( m x / a )x cos (nn/b)y etiOt-P) ...(3.59)

h'
H,, = %Jh C(mn/a) cos ( m d a )x sin (nn/b)y eti@-P)
.
...(3.60)

3.14. Various'TM Modes in Rectangular

Waveguides
For m = 0 and n = 0, all field components vanish. Therefore, TMoo
mode does not exist in a Rectangular waveguide.
For TMol mode, m = 0, n = 1. Here also all field components
vanish. Therefore, TMol mode too does not exist i n rectangular
waveguides.
For TMlo mode, rn = 1,n = 0. Again all field components vanish.
Therefore, TMlo mode too does not exist in rectangular wave guides.
For TMll mode, m = n = 1. All field components (Ex, Ey, Hx and
Hy)exist. They also exist for higher order modes.
2 2ab
Now AC - =$T
d(m/a12 + ( n / b )2 m"-T
bT+ n a
For TMll mode, Ac = 2ab
4Z'F
The cut-off wavelength for TM12and TMzl modes are given by,

kc (for TM12mode) = -2ab

-
4ZzF
2ab
kc (for TMzl mode) =
4x27
TRANSMISSION MEDIA 87

Of t h e t h r e e T M modes, T M l l mode h a s t h e highest cut-off

wavelength. Also, it will be higher than all other possible higher order
TM modes. Therefore, TMll mode is the dominant mode in rectangular
waveguides. The field pattern for TMll mode, the dominant Transverse
Magnetic mode, is shown in Fig. 3.29.

End View Side View

Lines of Electric Field -----Lines of MagneticField
0 Towards observer + Away from observer
Fig. 3.29
3.15. Power Loss in Rectangular Waveguides
While solving Maxwell's equations and the wave equations to ob-
tain expressions for different field components, i t was assumed that
the walls of the waveguide were made up of perfect conductors and that
the dielectric in the region between the four walls of the guide was
lossless. The propagation constant (y) under these ideal conditions is
given by :
r=w
The qyantity (h2)is a real number and a s mentioned earlier, its
value depends upon waveguide dimensions and the order of mode being
considered. In case of rectangular waveguides, h2 is given by

where a and b are broad and narrow dimensions of the waveguide.

Now r=a+jp
where a = Attenuation Constant
p = Phase Shift constant.
< h2
For operating frequencies less than the cut-off frequency, 0 2 p ~
and therefore (y) is real which implies that
Y=ol=&G&
When the operating frequency equals the cut-off frequency, y = 0 or
h = w f i . For all frequencies above cut-off, (y) is imaginary and the
attenuation constant (a)is zero. To sum up, for operating frequencies
below cut-off, (or) is a large number and all field components decay very
rapidly. For operating frequencies above cut-off, attenuation is practi-
cally negligible. All this is true for the ideal conditions where the
88 MICROWAVES AND RADAR

dielectric is considered as lossless and the waveguide walls are perfect-

ly conducting. Though the dielectric (air in case of hollow waveguides)
can be considered as nearly lossless, the walls of the waveguide do have
some loss due to finite conductivity of the material used.
When electric and magnetic fields suffer loss during propagation,
their magnitudes can be expressed by,
I E l = I E o z l e - " and I H I = I H o z J e - "
where EoZ ) and ] HoZ I are niagnitudes of field intensities a t z = 0.
The time average power flow would decrease proportionally to (e- 2") as
a function of ( 2 ) . If ( P ) is the total power a t the entry point to the
waveguide, (PT)the transmitted power and (PL)the power lost, then the
three are interrelated by,
P = P T + P L and P ~ = ( P ~ + P ~ ) ~ - ~ ~

PL/2
or Attenuation Factor = -
WT
Power lost per unit length
a=
2 x Power transmitted
The transmitted power can be obtained by integrating the axial com-
ponent of the Poynting vector over the cross-section of the waveguide.
Now axial component of Poynting vector, (P,) is given by :

For a lossless dielectric, the time average power flow through a

rectangular waveguide is given by :

where
For TEmnmodes, the average power transmitted through a rectan-
m l a r waveguide is Riven bv :
TRANSMISSION MEDIA 89

where 11 = W i s the intrinsic impedance in an unbounded dielectric.

In order to determine the power lost per unit wall area due to finite
conductivity of t h e wall material, t h e tangential magnetic field
strength computed for perfectly conducting walls is used to determine
the linear current density in the walls. I t is done so because tangential
magnetic field is expected to depend only slightly on the wall conduc-
tivity as long a s it is high. The square of the linear current density
multiplied by surface resistance of walls gives the power loss per unit
area in the walls. The power lost per unit length of the guide can then
be computed from :

where the integration is taken over the wall surface area of unit length
of the waveguide.
-
Rs= Surface Resistance = 4%
Substituting the values of (PL)and (PT)in the expression for (a),

3.16. Propagation of TE Waves in Circular

Waveguides
The procedure of determining various field components in case of cir-
cular waveguides is similar to the one followed in case of rectangular
guides. The co-ordinate system used in the present case is the cylindrical
co-ordinate system as it would simplify application of boundary conditions
and also it is going to be very convenient and helpful if Maxwell's equations
and the wave equations are expressed in cylindrical co-ordinate system.
Maxwell's equations can be expressed in cylindrical co-ordinates as :
90 MICROWAVES AND RADAR

...(3.72)

...(3.73)
For a TE wave propagating in z-direction, E, = 0, Hz + 0. Expanding
the wave equation v2HZ= - w2p3IZin cylindrical co-ordinates, we get,

-+
1 JH, +---
a2HZ ---- 1 d2H, +-=-w
d2HZ p&Hz
ap2 P a p p2 oq2 az2
$Hz 1 aH, 1 a2H,
or -+--+--+(~+w2p~)~z=~ ...(3.74)
ap2 P ap p2 a+2
Substituting + m 2 p = h2, we get
+-+>- ...(3.75)
ap2 P
p a~
Eqn. (3.75) is a partial differential equation having a solution of
the form
Hz = P.Q
where (P)and (Q)are respectively functions of (p) and (+) only.
Putting Hz = PQ in equation (3.741, we get

+)
a2(p~ 1 a2
(PQ)+ 7 -(PQ)+ I Z ~ P Q= 0 ...(3.76)
ap2 P ap P a+2
Since (PI is a function of ( p ) only and ( Q )is a function of ($) only,
equation (3.76)reduces to :

+BE+P&+~~~~=O
P ap p2
Multiplying both sides by (p2/p&),we get,
2 2
edP+~.1p+~&+,2~2=, ...(3.77)
P d p 2 P dp Q d$2
17
Let - d2& = - n 2 where (n2 ) is a constant, then
Q 4
2 2
~dP+~dP+(p2~2-~2)=0
P d p 2 P dP

or p 2 -+p-+(p
d2p dP 2h 2 - n 2) P = O
dp dP
'TRANSMISSIONMEDIA 91

Eqn. (3.78) is similar to Bessel's function of the form

Solution of equation of the type of (3.79) is given by :

Y = Cn Jn ( x )
where Jn(x) is nth order Bessel function of the first kind and (C,) is a
constant. Eqn. (3.78) may be rewritten as eqn. (3.80) to bring it to
standard Bessel form.

Eqn. (3.80) has a solution,

P = CnJn (ph)
Also, I d 2 =~- n2 ...(assumed earlier)
--
Q dq2
This has a solution given oy :
& = A n cosn$+Bn sinnq ...(3.82)
Fig. (3.30) shows the plot of Jn(ph), the nth order Bessel function
of the first kind as a function of (ph).
The complete solution for (Hz) can therefore be written as :
Hz=P.Q

= C ~ (ph)
J ~wzcos (nm+ t a n ' 21
.
= CnJn (ph) C,' cos n$

where C,' = and n$ = n$ + tan- 'A,/B,

Incorporating the z-variation factor, we get :
Hz = CnJn (ph) C,' cos n$ e-P
or Hz = CnCn' Jn(ph) cos n@-P
or Hz = Co Jn(ph) cos n+e-p ...(3.83)
where Co = C,.Cnt
Now, according to the boundary condition, tangential component of
electric field all along the surface of the circular guide is zero. If a is the
radius of the guide, then this boundary condition implies that E$ = 0 a t
p = a for ($) varying from 0 to 2n.

Therefore,
92 MICROWAVES AND RADAR

Using this condition in eqn. (3.83),

J,' (ah)= 0
where J,'(ah) denotes differentiation with respect to (ah).

Fig. 3.30
Now, it is the roots of J,' (ha)which need to be determined. These
roots for TEol, TEll, TEo2 and TE12modes are given by :
(ha)Ol' = 3.83, (ha)o2'= 7.02, (ha)ll' = 1.84 and (ha)lz' = 5.33
Various field components can be determined from :

Hz = C d , (ph)cos nqe-" ...(3.89)

Equations (3.84)to (3.88)have been obtained from set of equations
given earlier in Eqn. (3.68) to (3.73). The equations for various field
components of TE waves as determined from set of equations (3.84)to
(3.88)are as under.
TRANSMISSION MEDIA 93

H$ = j-2p Co
- J , (ph)sin n wp
h P

3.17. Propagation of TM Waves in Circular

Waveguides
For a T M wave, E, # 0 and Hz = 0 . The relevant wave equation to
be solved is:
V2E,=-w 2 W E , ...(3.94)
Following the same procedure a s followed earlier in case of T E
waves, eqn. (3.94)has a solution,
E, = Co J , ( p h )cos n @ e-" ...(3.95)
Applying the boundary condition, E, = 0 a t p = a , we get,
J,, ( p h )= 0
It is the roots of J,(ph) = 0 that need to be determined. These roots
for TMol, T M l l , TMo2 and TM12 modes are given by (ha)ol= 2.405,
(ha),, = 3.832, (ha)o2= 5.52 and (ha)l2= 7.106 respectively. Various
field components can be determined from :

Eqns. (3.96)to (3.99)have been ob&ned from eqns. (3.68)to (3.73).

The final expressions for various field co~rtponentsare then obtained by
substituting for (E,) in eqns. (3.96)to (3.9gj.These are :
-j C o w ~ n
H - -- J,, (ph)sin n $ C p
'- h2p
- jCooe
HQ=h J,' (ph)cos n$ e-v
94 MICROWAVES AND RADAR

E z = C oJ , (ph) cos n q e - y
Hz=0

3.18. Cut-off Wavelength in Circular Waveguides

For TE waves propagating in circular guide, the cut-off wavelength
(kc)is given by :

where (a) is internal r;Qm of the guide and (ha)' is solution of a Bessel
function. (h,) is maximum for minimum (ha)' . (ha)' is minimum (= 1.841)
for TEll mode which implies that TEll is the dominant TE mode in cir-
cular guides.
For TEll mode in circular guides (2m/1.841)
-I

For Transverse Magnetic (TM) modes in circular guides, the cut-off

wavelength (1,)is given by :

Again, (hc) has minimum valiic: (= 2.405) for TMol mode. There-
fore, TMol mode is the dominant 'fransverse Magnetic (TM) mode in
circular guides.
27142
hc (TMol mode) = -
2.495
Tables (3.1) and (3.2)list the characteristics of standard rectangular
and circular waveguides respectively.

3.19. Circular Waveguides

-Advantages and Disadvantages
The main advantages of circular waveguides include the following :
1.Circular waveguides are easier to manufacture.
2. Circular waveguides are easier to interconnect.
3. TEol and TMol modes in circular guides have rotational symmetry.
4. TEol mode in circular waveguides suffers lowest attenuation per
unit length of the guide and is therefore preferred for long distance
waveguide transmission. Theoretically, for TEo, modes in circular
waveguides, attenuation decreases indefinitely with increasing frequency.
TRANSMISSION MEDIA 95

The main disadvantages of circular guides include the following :

1.The plane of polarization rotates as the wave travels through the
circular guide due to roughness and discontinuities which in turn affects
the received signal.
2. Due to existence of infinite number of modes, the probability of
interference with the dominant mode is very much there.
3. For same cut-off wavelenglh, circular waveguides have a larger
cross-section and thus occupy more space as compared to rectangular
waveguides.
3.20. Power Loss in Circular Waveguides
The procedure for computing attenuation factor (a)in circular
guides is similar to the one outlined in case of rectangular guides.
Table 3.1. Characteristics of Standard Rectangular Waveguides
Physical dimensions Cutoff
%equency Recommended
EIA Inside, in cm fin.) Outside, in cm (in.) requency range
designa. for
-- air-filled for TElomode
tion
vaveguide in GHZ
Width Height Width Height in GHz

59.055 29.845 0.257 0.32 - 0.49

(23.250) (11.750)
53.973 27.305 0.281 0.35 - 0.53
(21.250) (10.750)
46.350 23.495 0.328 0.41 - 0.62
(18.250) (9.250)
38.735 19.685 0.394 0.49 - 0.75
(15.250) (7.750)
29.845 15.240 0.514 0.64 - 0.98
(11.750) (6.000:
25.400 13.018 0.606 0.76 - 1.15
(10.000) (5.125:
20.244 10.414 0.767 0.96 - 1.46
(7.970) (4.100:
16.916 8.661 0.909 1.14 - 1.73
(6.660) (3.410:
13.360 6.883 1.158 1.45 - 2.20
(5.260) (2.710'
11.328 5.867 1.373 1.72 - 2.61
(4.460) (2.310:
9.042 4.724 1.737 2.17 - 3.30
(3.560) (1.860:
7.620 3.810 2.079 2.60 - 3.95
(3.000) (1.500
6.142 3.233 2.579 3.22 - 4.90
(2.418) (1.273
5.080 2.540 3.155 3.94 - 5.99
-(2.000,
- (1.000

Table Contd.
96 MICROWAVES AND RADAR

Physical dimensions cutoff

'requency Recommended
EIA Inside, in cm (in.) h t s i d e , in cm (in.) *equencyrange
designa- for
air-filled fbr TElo mode
tion vaveguide in GHZ
Width Height Width Height in GHz
TRANSMISSION MEDIA 97

Table 3.2. Characteristics of Standard Circular Waveguides

--
W o f ffrequency for Recommended frequency
EIA 'nside diameter in
air-filled range for TErl mode in
designation cm (in.)
waveguide in GHz GHz
992
847
724
618
528
451
385
329
281
240
205
175
150
128
109
94
80
69
59
50
44
38
33
28
25
22
19
17
14
13
11
9

The attenuation factor (a)is again given by :

a = Power lost per unit length
2 x Power transmitted
98 MICROWAVES AND RADAR

Now, time average power transmitted through a circular guide can

be computed from :

Er E
where Z = -= - A = Wave impedance in the guide
H@ Hr
and a = radius of circular guide
Substitution of (2)for a particular mode in eqn. (3.107) yields the
power transmitted by that mode through the guide.
Now Z = n ....for TE modes
=q d m ....for TM modes
where 11 (also denoted by Zo) is intrinsic impedance in unbounded
dielectric. Eqn. (3.107) takes the form of eqns. (3.109) and (3.110) for
TE and TM modes respectively.

The power lost per unit length is determined in exactly the same
manner as outlined in case of rectangular waveguides in earlier pages.
Having determined power lost per unit of guide length and power
transmitted, attenuation factor (a)can then be computed.
- -

Problem 3.13 :A rectangular waveguide has dimensions 4 cm x 2

cm. Over what frequency range is such a guide single moded ?
Solution : The dominant mode (the mode with highest cut-off
wavelength) in rectangular waveguides is TElo mode.
The cut-off wavelength for TElo mode = 2a
where a = Broad dimension of waveguide = 4 cm
Therefore, cut-off wavelength = 8 cm
lolo- 3.76 DHz
Corresponding frequency = ------
8
The frequency range for single mode operation is the range of frequen-
cies corresponding to the dominant mode or the mode with highest cut-off
wavelength and the second highest cut-off wavelength. It is in this range
that only the mode corresponding to highest cut-off wavelength propagates
due to high pass filter type characteristic behaviour of a waveguide.
n
5'
Now cut-offwavelength in general is given by
.\l(m/a12 + (n/b12
TRANSMISSION MEDIA 99

The second highest cut-off wavelength mode here would be either

TEzo or TEol which respectively yield cut-offwavelength of (a) and (2b)
where (b) is the narrow dimension of waveguide. Since a = 2b in the given
problem, both yield a cut-offwavelength of 4 cm.This corresponds to a
cut-off frequency of 7.5 GHz.
Therefore, single mode operating range = 3.5 GHz to 7.5 GHz
In this range, only TElo mode propagates.

Problem 3.14 :An air filled rectangular waveguide has dimen-

sions 7.2 cm by 3.4 cm. Calculate group and phase velocities in the
dominant mode a t a frequency of 2.4 GHz.
Solution :Phase velocity (up) can be computed from :

where ;tg = Guide wavelength

h = Free space wavelength
c = Free space velocity of e.m. waves
Now,

h
Therefore,
%=-

This gives, -~=2~=6xlO~~cm/s

vP -
Also, if (vg) is the group velocity,

Problem 3.15 :Prove that (TMol) and (TMlo) modes do not exist
in a rectangular waveguide.
Solution :For a rectangular waveguide, the TM-mode expressions
for various field components are as follows :
-Y mk cos -
mx xsin 141E yetiot-w
~ X = s c ( a ) (a) (b)
(nl) ) [Y)
E , , = ~ c- sin - r cos - yetiw-'?)
100 MICROWAVES AND RADAR

It can be seen from these expressions that all the field components,
i.e. Ex, E,, Hx and H, vanish when either ( m )or ( n )is put equal to zero.
This implies that T M (where~ ~ m = 0 )and TMlo (where n = 0 )modes do
not exist in rectangular waveguides.

Problem 3.16 :Prove that for the propagation of dominant mode

in rectangular and circular waveguides, if the cut-off frequencies are
same, the cross-sectional area of the circular guide will be larger than
that of the rectangular wave guide. Make suitable assumptions.
Solution :Let ( a )and (b)be the broad and narrow dimensions of the
rectangular guide respectively and ( r ) be internal radius of the circular
guide.
Dominant mode in rectangular guide = TElo
Cut-off wavelength = 2a
Dominant mode in circular guide = TEll

For the two cut-off wavelengths to be equal,

2a = 3.41 r
a = 1.705 r
Now Area of cross-section of rectangular guide = a x b
Assuming a = 2b, which is a very reasonable assumption, we get
Area of cross-section of rectangular guide

Area of cross-section of circular guide = lcr2 = 3.14 r2

Ratio of the two cross-sectional areas

Therefore, circular guide is 2.16 times larger in cross-section as

compared to rectangular guide.

Problem 3.17 :Find all the possible modes that will propagate in
a rectangular waveguide having cross-sectional dimensions of 4 x 2 cm.
The operating frequency is 5 GHz.
Solution :Operating frequency = 5 GHz
Therefore, operating wavelength,
TRANSMISSION MEDIA 101

The necessary condition for a mode to propagate is that the cor-

responding cut-off wavelength should be greater than the operating
wavelength.
n
4
Now, the cut-off wavelength for TEol mode = for
d(m/a)2 + (n/b12
m=O,n=l
=2b=4cm
Cut-off wavelength for TE lo mode

=2a=8cm
Cut-off wavelength for TEll (or TMll) mode

Now, the operating wavelength is 6 cm.Only TElo mode has a cuboff

wavelength that is greater than the operating wavelength. Other modes
like TEol,TEll, TMll have cut-off wavelengths that are smaller than the
operating wavelength. Therefore, only TElo propagates in the given guide
at the given operating frequency.

Problem 3.18 : The dominant mode is propagating in a rectan-

gular waveguide of dimensions, a = 4 cm and b = 2 cm. Distance be-
tween the field maxima a n d minima is found to be equal to 4 cm with
the help of a travelling wave detector. Determine the frequency of the
wave.
Solution :The dominant mode in rectangular guide is TElo.
For TElo mode, cut-off wavelength
=2a=2x4=8cm
Distance between maxima and minima = 4 cm
This equals (hg/4) where (h) is guide wavelength.
Therefore, hg/4 = 4

Also, ko where h, is cutoff wavelength

kg =
1-( W A C )
102 MICROWAVES AND RADAR

Th:s gives fo = 7.1 55Hz = 4.192 GHz

lo1O

Problem 3.19 :A rectangular waveguide is chamterised by a = 6 cm,

b = 3 crn. If the operating wavelength is 4 cm, compute the Guide Wavelength,
Phase Sh@ constant and the Phuse Velocityfor the dominant &.
Solution :The dominant mode is TElo.
Cut-off wavelength for dominant mode, A,. = 2a = 2 x 6 = 12 cm
Guide wavelength,

Phase velocity, V p= (hg/?.)c = 4.24/4 c = 1.06 c = 3.18 x 10' m/s

Phase shiR constant, P can be computed from,

Problem 324) :A rectangular waveguide is filled with a dielectric

material of relative permittivity E, = 9. It has inside dimensions of 7 x 3.5
cm and operates in the dominant TElo mode. Compute the following :
1. Cut-off frequency
2. Phase velocity in the guide at a n operating frequency of 2 GHz
3. Guided wavelength at the same operating frequency.
Solution : The cut-off wavelength (A,.) for the dominant mode
=2a=14cm
C 3 x 10l0
Operating wavelength, h =

The guide wavelength,

-==a- -5cm

Phase velocity, vp =(A&) c = (5.35/5) c = 1.07 c = 3.21 x lo8 m/s

TRANSMISSION MEDIA 103

Review Questions
1. What are the major sources of losses in RF transmission
lines? To what extent it would be justified to assume RF
transmission lines to be lossless ?
2. Define Characteristic Impedance of a transmission line.
What would be the input impedance of a finite length of a
given line if it was terminated in its characteristic im-
pedance. Justify your answer.
3. Derive expressions for Attenuation Constant (a)and Phase
Shift Constant (P) for lossless as well as practical transmis-
sion lines.
4. The propagation constant of a line is found out to be purely
imaginary. What does this tell about the attenuation charac-
teristics of the line ?
5. Write short notes on
( a ) Smith Chart
( b ) Double Stub Tuner
( c ) Quater Wave Transformer
6. What do you understand by waveguide modes ? What are
dominant modes ?
7. Define the following waveguide parameters.
( a ) Cut-off wavelength
( b ) Group Velocity
(1.1 Characteristics Wave Impedance
8. Comment on the following statements.
(a) A waveguide acts like a high pass filter
( b ) TEM mode cannot exist in a hollow metallic waveguide
( c ) TErois the dominant mode is rectangular waveguide.
9. Briefly describe similarities and dissimilarities between two
wire transmission line and waveguides.
10. With the help of suitable expressions, explain how parameters
like Cut-off wavelength, Guide wavelength, Phase velocity
change in case of a rectangular waveguide filled with a dielectric
of relative permittivity (E,) as compared to a hollow waveguide.

Problems
1. A finite length of a transmission line having a characteristic
impedance of 50 f2 is terminated in 50 f2. Determine its input
impedance. Also determine the input impedance if the length
of the line is doubled. 150 Q, 60 Q]
104 MICROWAVES AND RADAR

2. Determine the input impedance of a half wave length long

section of a line having a characteristic impedance of 75 R
and terminated in 100 R. 1100 R1
3. A 15 cm long transmission line section is used a t a frequency
of 1 GHz. What would be the input impedance if this line is
terminated in 100 S2 ? Also determine the input impedance if
the same line is used a t 500 MHz ? The characteristic im-
pedance of the line is 50 R [I00 R, 25 Rl
4. For a 1cm x 2 cm rectangular waveguide, calculate the fol-
lowing :
( a ) Group velocity and phase velocity
( b ) Cut-off frequency
(c) Characteristic impedance
Operating frequency is 10 GHz.
L1.98 x 101° cm/s, 4.54 x 10" m/s, 7.5 GHz, 285 Rl
5. Calculate the characteristic wave impedance of a rectangular
waveguide, whose broader dimension is 7.5 cm, for TElomode
and an operating frequency of 6 GHz. [400Rl
6. The breadth of a rectangular waveguide is 10 cm. Calculate
its cut-off wavelength for the dominant mode. For a 5 GHz
signal to be propagated in this waveguide in the dominant
mode, calculate the guide wavelength, and the characteristic
wave impedance. [20 cm, 6.3 cm, 397 Rl
7. A waveguide h a s dimensions of 2 cm x 1 cm. Signals
propagate down the line a t 16 GHz and 20 GHz. For both
frequencies, determine the guide wavelength (i) in TElo mode
and (ii) in TEzo mode.
(i) hg = 2.1 cm at 16 GHz
hg = 1.6 cm at 20 GHz
(ii) hg = 5.4 cm at 16 GHz
hg = 2.3 cm at 20 GHz.
8. For the dominant mode to be propagated in an air filled
circular waveguide, the cut-off wavelength is 10 cm. Find the
required internal diameter of the guide. (5.86 cm)
SELF EVALUATION EXERCISE
Multichoice Questions
1. Velocity factor of a transmission line is 0.75.Electromagnetic
waves travel down the line with a velocity of
TRANSMISSION MEDIA 105

(a)3 x 10' m/s (b) 4 x 10' m/s

(c) 2.25 x 10' m/s (d) indeterminate from given data
2. Standing wave ratio (SWR) of unity implies that
(a) the line is terminated in its characteristic impedance
(b) the line is open circuited
(c) the line is terminated in a reactive load
( d ) the line is short circuited
3. Two transmission line sections ( x ) and Cy) cut from thd same
transmission line are 10 m and 20 m long respectively. Assuming
that characteristic impedance of ( x ) is 50 R, the same for @)
would be
(a) 25 R (b) 100R
(c) 50 R ( d ) 200 R
4. A transmission line when terminated in a certain load has a
reflection co-efficient of (1/3). VSWR is
(a) 3 (b) 2
(c) 113 (d) 4
5. If the characteristic impedance of the line in Q-4 is 50 R, the
load impedance would be
(a) 25 R (b) 100 R
(c) 200 R ( d ) 12.5
6. The SWR on a line is infinity when the line is open circuited
or short circuited or when it is terminated in
(a) its characteristic impedance
( 6 ) a reactive load
(c) a capacitive load with Q = 100
(d) an inductive load with Q = 100
7. Mark the correct expression :
(b) VSWR =1-IP(
l+lPl

8.A 75 R line is terminated in a load resistance of 100 R. Per-

centage of incident power reflected is
(a) 14% (b) 25%
(c) 10% (dl 2%
9. A quarter wave transmission line section is used to reject an
interfering frequency of 100 MHz. Its approximate length is
(a)3 m ( b ) 75 cm
(c) 1.5 m (dl 6 m
106 MICROWAVES AND RADAR

10. The co-axial cable belongs to

(a) TEM class of transmission lines
(b) quasi-TEM class of transmission lines
(c) non-TEM class of transmission lines
(dl none of these
11.The characteristic impedance of the transmission line with
the equivalent circuit a s shown in Fig. 3.31 will be
(a) 50 R (b) 2500 R
( c ) 100 SZ (dl 75 R
10 mH 10 mH 10 m H
------- -------

1C mH 10 mH 10 mH
Fig. 3.31.
12. A quarter wave transmission line section is used to match a
75 R line to a 300 R load. The matching section should be cut
from a line having a characteristic impedance of
(a)75 R (b) 300 R
(c) 150 R (dl 50 R
13. The characteristic impedance of a transmission line is
(a)directly proportional to its length
(b) inversely proportional to its length
(c) independent of its length
(dl directly proportional to square root of its length
14.The input impedance of a 75 R line terminated in a 300 SZ
load will be (assume the line to be h / 2 long)
(a) 300 R (b) 150 R
(c) 75 R (dl indeterminate from given data
15. A transmission line that is (U2)long a t the operating fi-equency
is terminated in a load resistance of 50 R. Its input impedance
is
(a) indeterminate from given data
(b) 50 R (c) 100 R (d) 25 R
16. A given transmission line when open circuited a t the load end
has an input impedance of (Z1) and when short circuited, it has
-
an impedance of (Z2).The characteristic impedance of the h e is
TRANSMISSION MEDIA 107

17. A sh& circuited (U4)line has an input impedance of

(a) zero ( b ) infinity
( c ) 50R (dl none of these
i50R is the characteristic impedance of the line)
18. If R, L, G and C are respectively the resistance, inductance,
conductance and capacitance per unit length of the transmis-
sion line, then the Heaviside condition for the distortionless
transmission line would be
( a ) R/C = L / G (b)RL=CG
( c ) RC = LG ( d ) RG = LC
19. The distance between adjacent minima and maxima of a
standing wave on a transmission line is
(a) h (b) h/4
(c) h/2 (dl none of these
20. The length of a trammission line connected from the output
of the transmitter to the antenna is 10 m. The length of
another trammission line connected from the output of the
same transmitter to another antenna is 10.5 m. If the signal
frequency is 100 MHz, the signals reaching the two antennas
shall have a phase difference of
( a ) 0" ( b ) 30'
(c) 60' (dl 120"
21. A lossless transmission line is excited by a 200 MHz signal.
If the dielectric constant is 2.25, the phase shift constant is
( a ) 2x rad/m ( b ) x rad/m
( c ) 1.5 rad/m (dl 3 radlm
22. A lossless transmission line is terminated in a mismatched
load. If the normalised load impedance is (3 +j3) Q, the reflec-
tion co-efficient is
( a ) 0.66 ( b ) 0.75
(c) 0.8 ( d ) indeterminate from given data
23. Whenever a lossless transmission line is terminated in a
resistive load impedance larger than the characteristic im-
pedance of the line, the first maximum of the voltage standing
wave pattern starting from the load end occurs at
( a ) ( h / 4 )from the load end
( b ) ( h / 2 )from load end
( c ) Load end itself
( d ) anywhere between ( U 4 )and ( h / 2 )from load end
24. Higher the radius of constant SWR circle on Smith chart
(a)lower is the impedance mismatch
( b ) higher is the magnitude of reflection co-efficient
( c ) higher is the permittivity of transmission line dielectric
(dl lower is the propagation velocity
25. A 300 R lossless transmission line was analysed using Smith
chart. The input impedance a t a certain point on the line as
observed from the constant SWR circle was represented by a
certain point 'P. Then the point on the SWR circle located
diagonally opposite to the point 'P'represents.
( a ) reflection co-efficient
( b ) input impedance only
( c ) input admittance only
(dl none of these
26. A waveguide can be considered to be equivalent to a
( a ) low pass filter ( b ) high pass filter
( c ) band pass filter ( d ) band reject filter
27. One of the following modes does not exist in waveguides
( a ) TEM ( b ) TEll
( c ) TEIO (dl TMol
28. For a given rectangular waveguide, the cut-off frequency for
TElo mode is always
( a ) higher than that for the TEll mode
(b) lower than that for the TEll mode
(c) equal to that for the TEll mode
( d ) 100 MHz
29. The dominant mode in rectangular waveguides is
( a ) TElo ( b ) TEll
( c ) TMol (dl TMll
30. Ir a rectangular waveguide with wide and narrow dimen-
sions as ( a ) and ( b ) respectively, the cut-off wavelength (&)
for TE,, modes is given by

31. The wider and narrower dimensions of a rectangular waveguide

are 2 cm and 1crn respectively. For the TElo mode to propagate,
the signal frequency
( a ) should be larger than 7.5 GHz
( b ) should be larger than 3.75 GHz
(c) should be smaller than 7.5 GHz
(dl should be larger than 15 GHz
TRANSMISSION MEDIA 109

32. The characteristic impedance of a n air dielectric rectangular

waveguide mode from a non-magnetic material mainly
depends upon
(a) the cut-off wavelength
( b ) the guide wavelength
(c) the dimensions (narrow and wide) of the waveguide
( d ) the propagating mode
33. If (h)is the free space wavelength, (LC)is the cut-off wavelength,
then the guide wavelength (Ag) is given by :

(c' h-g = ho (d) none of these

41- (who)2
34.The phase velocity (up) and group velocity (vg) are interrelated
by :
(a)Vg VP = c2 (b) Vg/Vp = c

35. When the free space wavelength equals the cut-off wavelength,
(a) the group velocity equals phase velocity
(b) phase velocity becomes zero
(c) group velocity becomes zero
(dl group velocity becomes infinite
36. A 10 GHz wave is propagating in a waveguide having a wall
separation of 4 cm. The largest number of half waves of
electric intensity possible in this waveguide is
(a) 1 (b)2
(c) 3 (d) 4
37. If (Zn)is the characteristic wave impedance of free space, then
the characteristic wave impedance (2) of a waveguide for
TE,, is given by

20
(c) Z = d (d)~none of these
1 - (Ac/Lg)
38. Refer to the rectangular waveguide shown in Fig. 3.32(a).The
cut-off wavelength for this waveguide is given by
110 MICROWAVES AND RADAR

2
( a ) kc = -
d ( n / a ) 2+ (n/b12

39. For the waveguide shown in Fig. 3.32(b),the cut-off wavelength

for TE modes is
(a) 4 cm ( b ) 1.414 cm
( c ) 2 cm ( d ) 4 6cm
40. For a lossless rectangular waveguide, the operating wavelength
is 0.6 times the cut-off wavelength for the operating mode that
is transverse magnetic in nature. The wave impedance in the
waveguide is
( a ) 1 2 0 ~ohm ( b ) 93 x ohms
(c) 150 x ohm ( d ) 11one of these
Transmission of a wave in a closed waveguide is possible only
when
( a ) the frequency of operation is less than a certain critical value
( b ) the phase shift (P) is real and positive
(c) the phase shift is zerc
( d ) none of these

(a) Fig. 3.32. (b)

42. A rectangular waveguide has dimensions of ( a ) a n d ( b )

respectively along x and y axes, z-axis being the direction of
propagation. Then phase shift constant (P) is
(a) - ( m ~ / a-)( n~ a / b )2 (b) m 2 ~
( c ) m 2 p + ( r n ~ / a+)(~n ~ / b ) ~ (dl m 2 / 6
TRANSMISSION MEDIA 111

43. The most dominant transverse electric mode in an evacuated

circular waveguide is
( a ) TE lo mode ( b ) TEol mode
( c ) TEll mode ( d ) TEzomode

44. Fig. 3.33 shows the longitudinal and cross-sectional views of

a certain mode in a n evacuated rectangular waveguide.
Guess the type of inode.
( a ) I t is not possible to guess unless other views are also given
( b ) I t is TE lo mode
( c ) I t is TMIOmode
(dl It is TEol mode
45. TEol mode exists in a n evacuated circular waveguide of
radius 2 cm. Its cut-off wavelength is
(a)indeterminate from given data
( b ) 3.28 cm ( c ) 2 cm ( d ) 4 cm

Longitudinal Cross-sectional
Fig. 3.33.

Match the following

Match the transmission line sections shown on the left with the
corresponding equivalent circuits on the right.
Transmission line section Equwalent circuit

Fig. 3.34.
 Microwave Components
t microwaves, it is more appropriate and convenient to talk in

A terms of electric and magnetic fields propagating in the trans-

mission medium rather than voltages and currents that we are
familiar with in case of transmission lines. A waveguide
provides a medium for the guided propagation of electric and magnetic
fields at microwave frequencies. This most con~n~only used transmis-
sion medium was discussed in detail in chapter-3. There is a large
variety of passive microwave conlponents that are either relevant to the
use of waveguides or are derived from waveguides and that are indis-
pensable for any microwave system. These include tools such as Cou-
pling Probes and Loops used to couple the microwave signals to the
waveguides, Waveguide Junctions or Tees used to either combine or split
two or more nlicrowave signals in a waveguide system, Irises! Windows
and Posts used to provide impedance matching, Cavity Resonators used
as tuned circuits, Directional Couplers mainly used for unidirectional
power flow measurement, Isolators and Circulators, Phase Shifters and
so on. Other passive components very relevant to nlicrowave systems
are the Striplines, Microstrips and SAW devices. Striplines and
Microstrips are also transnlission media developed as an alternative to
waveguides for applications that require nliniaturisation. SAW devices
make use of propagation of surface acoustic waves on the surface of a
solid piezoelectric material and are used as delay lines, filters, resonant
circuits etc. All these components would be discussed at length in the
present chapter as regards their physical appearance where-ever im-
portant, principle of operation, properties and applications.
The topics co-~eredin this chapter include :
4.1. Coupling probes a n d 4.9. Attenuators
loops 4.10. Phase shifters
4.2. Windows 4.1 1. Cavity resonators
4.3. Tuning posts a n d screws 4.12. Wavemeter
4.4. Waveguide junctions 4.13. Hybrid ring
4.5. Directional couplers 4.14. Slotted line
4.6. Isolators a n d circulators 4.15. Striplines a n d
4.7. Waveguide flanges Microstrips
4.8. Rotating joints 4.16. SAW devices
MICROWAVE COMPONENTS 113

4.1. Coupling Probes and Loops

Probe Coupling and Loop Coupling are the two commonly used
techniques for coupling microwave signal to the waveguides. Both probe
and loop coupling can be used to launch a particular mode in a
waveguide. While probes couple primarily to an electric field, loops
couple to a magnetic field. But i t must always be borne in mind that the
two fields are inseparable and in case of both probe and loop coupling,
both an electric as well as a magnetic field will be set up.
Coupling Probes
A coaxial line may be coupled to a waveguide by means of either a
probe parallel to the electric field a t or near the point where electric
field has maximum value.

Wavegu~de \
Electr~cfield
Fig. 4.1
A coupling probe usually consists of s n extended inner conductor
of a coaxial cable a t the mid-point of one of the wide walls i.e. one of
walls normal to E field pattern as shown in Fig. 4.1. The figure shows
the progress of set of two lines of electric field a t a point in the wave as
the wave passes through a junction of coaxial line and waveguide. Here,
higher order modes are also excited because the electric field in the
vicinity of the probe has components normal to the axis of the probe.
Also both the electric and magnetic fields in the vicinity of probe differ
in other respects from the desired TElo mode. With the proper choice
of waveguide dimensions, i t is possible to suppress higher order modes
to a great extent within a wavelength or two. Generally, a short circuit
terminates the waveguide and the probe is placed approximately a
quarter wavelength from the termination. When nrobe is used for
coupling out microwave power, i t is a t an antinode of electric field. On
the other hand, when i t to be excited, direct waves from probe are
reinforced by waves reflected from closed end of the guide.
For nlininlisation of reflections at the junction, the probe must be
matched to the waveguide. Proper choice of probe length and position
of probe relative to closed end of the guide helps ,accomplishing the
matching. The length and position can be best obtained experimentally.
114 MICROWAVES AND RADAR
Flaring and rounding off the end of the probe helps obtain satisfactory
matching over a large frequency range. Adjustable matching elements
have to be used to obtain close matching. Analogous to impedance
matching, two adjustments have to be done in order that both mag-
nitude and phase are matched. One adjustment can be obtained by
replacing the fixed closed end of the waveguide by a movable short-cir-
cuiting plunger, and the other adjustment can be achieved by use of
stubs.
Short probes af.the types shown in Fig. 4.1. are used to measure
the field strength within the waveguide without coupling out much
power. Fig,X2 shows a modification of the coupling probe in which the
central fonduckor of coaxiable cable extends fully across the waveguide
and terminates on the distant wall. This has the advantage that the
end of the centre conductor is rigidly supported. For impedance match-
ing, the centre conductor may pass through the distant wall and
terminated in a coaxial line stub as shown in Fig. 4.3.

Fig. 4.2

-Energy

Fig. 4.3
Matching over an appreciable frequency band with a fixed system
gets complicated by the inductive reactance of the extended centre
conductor of coaxial cable and the shunting susceptance of the short
circuited section of the waveguide beyond the junction at frequencies
different from that at which this section is L/4 long. The effect of the
MICROWAVE COMPONENTS 115
centre conductor reactance may be reduced by malung tbe ratio of
diameter of centre conductor to width of waveguide an optimum value.
Fig. 4.4 shows the photograph of a typical waveguide adaptet that
provides loosless transition from an N-type coaxial line to a rectangwlar
waveguide.

r-

Fig. 4.4
Coupling loops
In a number of microwave systems, it is often desirable to couple
a coaxial line to a waveguide or a cavityresonator by means of a coupling
loop rather than by a coupling probe.
Coupling

( Equivalent)

Fig. 4.5
116 MICROWAVES AND RADAR
Fig. 4.5 shows the use of a coupling loop in s junction between
coaxial line and waveguide. The loop couplicg ic basically magnetic, so
the loop must be placed at or near the pcini of maximum magnetic field
strength and turned in such a direction that its plane is normal to flux
lines. The loop can be mounted at the end of the shorted waveguide or
in the middle of top or bottom wall a t a distance n. h / 2 from the short
circuited end where n is an integer and (A)is the wavelength.
The plane of the loop should be perpendicular to H field for
maximum coupling. The degree of coupling obtained with a loop
depends upon its size, shape and orientation ;and in general increases
with the area of loop. Increasing the size of loop increases self-induc-
tance which is not desirable. When length of the loop conductor is
appreciable in comparison to wavelength, then, there will be appreci-
able potential gradient along the conductor and magnetic coupling may
be accomplished, accompanied by appreciable electric coupling too.
Mechanical, as well as electrical considerations have to be taken
into account while making a choice of coupling probe or loop. The
important factors to be considered are :
(1) Possibility of voltage break down in the vicinity of an
antinode
(2) Ease of adjusting the coupling
(3) Constancy of coupling as mechanical changes are made
(4) Prevention of interference with the electron stream
(5) Matching
For example, the loop extending from a movable short circuiting
wall of a rectangular waveguide supporting the TElo mode is always at
a point of maximum magnetic field, regardless of position of the wall.
The loops find use in microwave oscillators in preference to cou-
pling probes because of the fact that a probe in the proper position for
adequate coupling would interfere with electron movement within the
tube.
4.2. Windows
Waveguide Windows, also known as Diaphrams, Apertures or
Irises, are used to provide impedance matching in the waveguides in
the same way as we used stubs in case of transmission lines. Three
common types of windows include :
1. Inductive Windows
2 . Capacitive windows
3. Resonant Windows
Inductive Windows
Conducting diaphragms extending in a waveguide from side walls
a s shown in Fig. 4.6 have the effect of adding an inductive susceptance
MICROWAVE COMPONENTS 117
across the waveguide at the point at which diaphragms are placed. This
is because, the iris in Fig. 4.6 allows current to flow where none flowed
before. The electric field that advanced
before now has a conducting surface in its
plane, which permits current flow. Thus Project~onfrom
some energyis storedin the magnetic field
which leads to an increased inductance at
that point of the waveguide. Such an ele-
ment is therefore, called an inductive win-
dow. The amount of normalised inductive
susceptance added is a function of the Fig.4.6
window insertion distance I.
Capacitive Windows
Conducting diaphragms extending into the waveguide from top
and bottom walls constitute what is known as a capacitive window as
shown in Fig. 4.7. These windows produce

Wavegu~de
walls

Fig. 4.7
the effect of a capacitive susceptance shunted across the waveguide at
that point. It is obvious that the potential which earlier had existed
between top and bottom walls ofwaveguidenow exists between surfaces
that are closer. This results in an increased capacitance at that point.
Capacitive windows are not used extensively because of the danger of
voltage breakdown which ultimately places a limit on the power that
can be transmitted through the waveguide.
Resonant Windows
A conducting diaphragm of the form shown in Fig. 4.8 (a)gives the
effect of a parallel tuned LC circuit connected across the guide at the

(a) (b)
Fig. 4.8
118 MICROWAVES AND RADAR
point where diaphragm is placed. An equivalent circuit is shown in Fig.
4.8(b).As a first approximation, a resonant window may be considered
to be a combinationof an inductive and a capacitive window, at the same
point in the guide. If the inner dimensions of aperture are properly
chosen, the frequency range covered is large. However, a limit of
minimum aperture size prevents any further changes. The value of Q
that can be obtained is of the order of 10 and decreases as the size of
aperture is increased. Since impedance offered by the resonant window
is very high for the dominant mode, and the shunting effect is negligible
for the same mode, other modes will be significantly attenuated.
Windows are usually employed only to correct a permanent mis-
match, rather than to provide adjustable matching.
4.3. Tuning Posts and Screws
Tuning Posts and screws are also used for impedance matching. A
cylindrical post, extending into the waveguide from one of the broad
sides has the same effect as a window in providinglumped capacitivelin-
ductive reactance at that point (Fig. 4.9).

1
Capacitive, Inductive Waveguide

Fig. 4.9
When a cylindrical post extends only slightly into the waveguide,
a capacitive susceptance is provided at that point. However, when the
depth of penetration is h/4, a series resonance occurs and if insertion
is greater than h/4, an inductive susceptance results at that point,
which decreases as insertion is more complete. The sharpness of
resonance at mid-point insertion is an inverse function of diameter of
post. This can, therefore, be used as a band stop filter, to allow the
propagation of higher mode in purer form. The advantage of such posts
over windows or irises is the adjustability they provide, which results
in case of matching.

b Waveguide

Fig. 4.10
A tuning screw as shown in Fig. 4.10, is a n o t h e way to obtain
variable susceptance at the desired point in the waveguide. Again, the
reactance offered by the tuning screw at the point of insertion in the
waveguide is capacitive for insertion less than (h/4) and inductive for
MICROWAVE COMPONENTS 119

insertion more than (h/4). It is a series resonance circuit for (h/4)

insertion. The narrow dimension of waveguide is ordinarily too small
to allow the use of screws to produce a wide range of inductive suscep-
tance.
To reduce power leakage, half wavelength transmission line
chokes may be frequently incorporated in the screw-assembly. How-
ever, the most direct method of matching is to use a single screw tuner
which can be adjusted both in length and position, thus becoming
analogous to the single stub matching in transmission lines. Usually
matching is done by trial and error till the VSWR is in the allowable
range.

4.4. Waveguide Junctions

Wageguide Junctions are used whenever it is desired to combine
two or more signals into one or split a signal into two or more com-
ponents in a waveguide system. The commonly used waveguide junc-
tions include :
(a) E-plane tee junction
(6) H-plane tee junction
( c ) Hybrid junction

E-plane Tee
E-plane tee (Fig.4.11) is a voltage or series junction, symmetrical
about the central arm, so that the signal to be split up is fed from it or
signals to be combined are taken from it. However, the problem has
more complexities than it appears superficially. This is because some
form of unbanted reflections occur and it is essential to provide some
sort of impedance matching to minimise reflections. In-&&, E-plane
tees may themselves be used for impedance matching purpose in a
manner similar to the short circuited transmission line stub.

Fig. 4.11
120 MICROWAVES AND RADAR
Fig. 4.12shows the propagation of an E-field through the E-plane
tee junction when electromagnetic waves in TElo mode enter the
junction from side arm.

Rending of the field as it leaves the side arm causes fields of

opposite sense to be set up in the main arm. If the j~nctionis completely
symmetrical, the waves of opposite phase that leave through the main
arms are equal in lhagnitude (Fig.4.12a). The manner in which E-field
(TElo) behaves when waves enter the junction through main arms is
shown in Rg. 4.12 (b). Rending of lines causes a portion of E-field to
appear across one enterance of the side arm. If the instantaneous field
entering the main arrns from opposite directions are equal in magnitude
and phase the net field in side arm is zero.
In general, the resultant fieldleaving the junction through the side
arm is proportional to the difference between instantaneous fields that
entered the junction from opposite directions. Also, the effective value
of the field leaving through the side arm is proportional to the phasor
differencebetween tho ontering fields. Maximum energy leaves the side
arm when waves pntering the junction through two main arms are in
phase opposition.
H-plane tee
H-plane tee is so called because the axis of the side arm is parallel
to the planes of the H-field of the main transmission line. As all three
arms of H-plane tee lie in the plane of magnetic field, the magnetic field
divides itself into the arms. This is thus a current junction.
If the H-plane junction as shown in Fig. 4.13 is completely sym-
metrical and waves enter through the side arm, the waves that leave
through the main arrns are equal in magnitude and phase. Since the
electricfield is not bent as the wave passes through an H-planejunction,
but merely divides between two arms; fields of same polarity approach-
ing the junction from the two main arms produce components of electric
field that add in side arm. The effective value of field leaving through
MICROWAVE COMPONENTS 121
the side arm is proportional to phasor sum 01 fields entering the
junction.

(a) (b)
Fig. 4.13
Maximum onergy delivery to side arm occurs when waves entering
the junction through main arms are in phase. The standingwave in the
main line then has an antinode of electric field at the junction, and a
current-node at the same jcnction. High energy delivery to a branch
line connected to a transmissi~~n line at a point of high voltage and low
current takes place if branch line is connected in shunt with the main
line.
If the side branch of a senes or parallel tee has negligible loss and
is terminated in a short circuit or an open circuit, reactance is then
inserted in series or shunt with the line. The magnitude and type of
reactance (positive or negative) may be readily adjusted if the length of
short circuited side branch is adjusted by means of a plunger.
Hybrid unctions
The most commonly used hybrid junction is the Magic Tee. A magic
tee is a combination of an E-plane tee and an H-plane tee. It acts as a
4-port hybrid circuit, which in general form is shown in Fig. 4.14.

Fig.4.14
The characteristics of the hybrid circuit are such that if power
enters the circuit through arm A or C ;the power is delivered entirely
to arms B and D, with no power transmission from portA to port C or
C toA. Also power entering through armB or armDis deliveredentirely
to arms A and C, with no direct transmission from B to D.
122 MICROWAVES AND RADAR

Fig. 4.15.

-
Fie. 4.16
Fig. 4.15 shows bow a combination of an E-plane tee and an
H-plane tee makes a magic tee. Fig. 4.16 shows photograph of hybrid
junctions of various sizes and ahapes. The waveguide dimensions are
such that only TElo mode is supported. E-field in parallel arm P is
MICROWAVE COMPONENTS 123
normal to that in series a r m s ;therefore, th&e is no direct transnlission
between series and parallel arms, if they are symmetrically placed. The
characteristics of series and shunt t es are such that when waves of
\
equal amplitude and phase enter the and S arms, the E-fields cancel
in one of the side arms and add in the other as shown in Fig. 4.17(a).

Fig. 4.17
The energy applied to arm P or S is divided equally between 1and
2 arms, none emerging from opposite arm. When power enters through
arm 1and 2, the fields leaving the P arm are proportional to the phasor
sum of two input fields [Fig. 4.17 (d)],while the fields leaving arms S
are propurtional to phasor difference of two input fields, Fig. 4.17 ( e ) .
If arms 1and 2 are terminated in matched loads and no reflections
take place inside the junction, enterance of power through either arms
S or P results in equal power delivery to arm 1and 2. Reflections may
take place due to severe discontinuities in the junction and because at
the junction each arm is effectively terminated by two other arms of
equal impedance in series or parallel.
Effect of Reflections : (a)Only a portion of the power that
approaches the junction through arm S or P is delivered to arms 1and
2.
( b )Power is not delivered equally between arm 1 and 2 when power
enters thfough arm S or P.
(c) Balance does not exist between arms 1 and 2, that is, some
power transmits directly from arm 1to arm 2 or from arm 2 to arm 1.
Reflections must, therefore, be avoided or compensated for.
Applications of Magic Tee :A magic tee is commonly used as a
matching device, an isolator, a phase shifter and a T/R switch.
124 MICROWAVES AND RADAR
Isolator :Fig. 4.18 shows the use of magic tee as an isolator. As
shown in Fig. 4.18(a)the energy transmitted from the transmitter Tx
is divided between the matched load and the antenna but none goes to

Matched
load
Matched
load
Local

(a) (b)
Fig.4.18
the receiver Rx. On the other hand, energy received by antenna is
however delivered largely to receiver. This system is not often used
since one half of the transmitted power is sacrificedin the load resistor.
In Fig. 4.18(b),half of local oscillator power is delivered to frequen-
cy converter in the receiver Rx, but none of it reaches the antenna. On
the other hand, energy from antenna is divided between receiver and
matched load.
Matching Device :A matched magic tee, the arms of which are
terminated in matched loads, has a useful characteristic when it is fed
simultaneously through both members of either balanced pairs of arms
P and S or arms 1 and 2, from sources of same frequency. When power
enters through arms 1 and 2, fields leaving arm P are proportional to
phasor sum of two input fields, and that leaving a r m s are proportional
to difference of phasor fields entering the junction.
Adjustable Phase Shifter :Another application of magic tee is
as an adjustable phase shifter. Power is fed into the unit through arm
P and delivered through a r m s . As shown in Fig. 4.19,arms (1)and (2)
contain movable short circuiting plungers, one of which is h/4 farther
from the plane of symmetry than the other one. The input power from
arm P divides equally at the junction between arms (1)and (2); where
it gets reflected and returns to the junction 180" out of phase because
of short-circuiting plungers.

circuiting plungers

Fig.4.19
MICROWAVE COMPONENTS 125

Since the output of S arm is the phasor difference of inputs through

(1)and ( 2 ) arms, and output of P arm is the phasor sum of inputs
through arms (1)and (2) ; no output appears from P and all energy is
delivered to arm S. The phase of the output relative to the input
depends upon the positiox of plungers.
To sum up, we can make the following observations with reference
to the hybrid junctions :
(1)It is impossible to match a Tee-junction perfectly.
(2) I t is always possible to place a short circuit in one port of a
Tee junction in such a position that the remaining two ports
become electrically isolated. Two ports are said to be electri-
cally isolated when power flowing into the network in one
part is not transmitted to the other.
(3) A short circuit in the port of symmetry of a symmetrical
Tee-junction can be so placed that perfect transmission be-
'tween other two arms becomes possible.
4.5. Directional Couplers
Directional Couplers are mainly used for unidirectional power flow
measurement and SWR measurement.
A directional coupler (Fig. 4.20) is a junction between 4 pairs of
terminals having such characteristics that there is free transfer of
power without reflection between terminals C and D and no transfer of
power between terminals A and C or between terminals B and D. Degree
of coupling between A and D and between B and C depends upon the
structure of unit.
The important performance parameters of a directional coupler are :
( a ) Coupling Factor or Coefficient ( b ) Directivity
( c ) Isolation ( d ) Insertion Loss
( e ) Bandwidth ( f ) Frequency sensitivity

Fig. 4.20 .
Couplingr Factor :The Coupling Factor of a directional coupler is
the ratio of the input power to the coupled output power, expressed in
decibels. Thus, if power out of arm B is one-hundredth of the power into
arm C, the component is a 20 dB coupler. It should be noted that the
power out of arm D is reduced by the amount coupled out a t arm B. In
the 20-dB coupler, since only one per cent of power has gone to arm B,
the power out of arm D must be 99% of the input. 3y the same logic it
126 MICROWAVES AND RADAR

is seen that in the case of a 10 dB coupler, power output from arm D is

only 90% of the input ;while it is only 50% in the case of a 3 dB coupler.
PB
In general, coupling factor = - 10 loglo - dB
PC
where PB is power delivered to a matched load a t terminal B when
power PC enters through terminal C with matched loads a t D and A.
Coupling factor is a measure of energy levels in primary and secondary
wave guides. Thus, if coupling factor is known, small power observed
at B can be used to obtain total power in primary guide.
Directivity : Directivity is defined a s the ratio expressed in
decibels of the power out of the coupling arm to the unwanted signal in
arm A. For example, if the signal, in arm A is just one per cent of the
coupled signal, the directivity is 20 decibels. Note that the directivity
can be more or less than the coupling ; a 10 dB coupler with 30 dB
directivity would be one in which power out of coupled arm was 10 dB
down from the incident, and power out of isolated arm was 40 dB down
from the incident, or 30 dB below the coupled signals.
Directivity = 10 log PB/PA ; where PB and PA are powers delivered
in matched loads a t terminals B and A, wich a matched load a t D.
Directivity is a measure of how well the forward travelling wave in
primary guide couples only to desired terminals of a secondary wave
guide. Ideally, directivity should be infinite ;but well designed couplers
have a directivity of 30 to 35 dB.
Isolation :Isolation is a measure of how far the directional coupler
deviates from an ideal design. The isolated port (the port other than the
coupled port in the secondary line) receives very little microwave ener-
gy. The microwave energy received by this port is in reality a measure
of the leakage present a t this port. Isolation (in dB) can be computed
from :
Isolation (in dB) = Directivity (in dB) + Coupling (in dB)

F:]
Also, Isolation (in dB) with reference to Fig. 4.20 = 10 log -

Insertion Loss : The term 'Insertion Loss' has the same sig-
nificance with respect n directional couplers as for other components
in a microwave system. That is, it describes the loss resulting from the
insertion of the device into a transmission system.
Bandwidth :Bandwidth is the range of frequencies within which
the performance, with respect to some characteristic. falls within
specific limits.
Frequency Sensitivity :The maximum peak to peak variation in
the coupling factor that may be expected over a specified frequency
band is called the Frequency Sensitivity.
MICROWAVE COMPONENTS 127

In microwave circuits, usually only three of Ihe four ports are used ;
the unwanted port usually being terminated by a matched load built into
it. The built-in load must have an excellent VSWR, since any reflection
from it will appear almost entirely at the coupling arm.This directional
characteristic is achieved by providing two or more coupling mechanisms
between the two transmission lines. The couplings are so arranged that in
one direction of propagation in the secondary line, all coupled signals add
in phase ;while in the other direction their phases and amplitudes exactly
cancel. A simple example is a two hole directional coupler described below.
Two-hole Directional Coupler : The two hole directional
coupler is a four port component as shown in Fig. 4.21. Two holes P and
Q are located inside the coupler. The holes P, Q are spaced h,/4 apart. As
there is flow of signal energy from port (1)to (2), there is s r k e diversion
of energy into holes P and Q. If diameters of P and Q are same, then the
amount of energy through P and Q will be same. At P and Q, the energy
is further bifurcated into two portions -going towards port (3)and port (4).
Now the phase difference between the portions of energy going from P and
Q towards port (4) is zero, hence the two signal components add up ;
whereas the components having the direction towards port (3) are out of
21c 41c
phase by - x 2 x PQ = - x -g = 1c radians. Hence they cancel each other
kg 1,. 4
if their magnitudes are same, and this is the reason for absence of signal
at port (3).The number of holes may be more than 2, but it has to be an
even number of holes. In general, greater the number of coupling
mechanisms, the wider the frequency bandwidth and greater the direc-
tivity.

Fig. 4.21.

t
CJl
a : w ~ d t hof wavegu~de
d : d ~ aof holes
G
-
Q
2
0
U

d/a
Fig. 4.22.
- Fig. 4.23.
d-
 128 MICROWAVES AND RADAR

Fig.4.24.
Fig. 4.22 shows the graph of coupling factor a s a function of ratio
of hole diameter ( d ) to the waveguide% wide dimension (a).Fig. 4.23
shows the graph of coupling factor v x s u s hole diameter (dl for dif-
ferent ratios of ( b l a ) .( a )and ( b )are wide and narrow dimensions ofthe
wavequide.
Fig. 4.24 shows photograph of a tvpical waveguide directional
coupler.
Loop Directional Coupler :Pxother type of combined coupling
directional coupler is the capacitarce loop directional coupler shown in
Fig. 4.25. Here the loop D is exterding into the main line such that it
is getting influenced both by electric a s well a s magnetic fields.

c i
(coax~alcable)

Fig. 4.25
When a n electric field acts on loop D, there is a charge generated
. and it behaves a s if there is a voltage source between loop D and the
main line M ; because of the capacitance between loop D and main line
M. [Fig. 4.26 ( a ) ] Hence
. we see that a n electric current I, flows in up-
ward direction in A and B arms of loop D, the magnitude being equal
in two arms.
MICROWAVE COMPONENTS 129

Now the magnetic field also causes a n auxiliary source to exist and
current produced due to this auxiliary source of magnetic effect is op-
posite to that in main line Fig. 4.25 ( b l . Hence we have two currents :
(i) Electric current I,. and (ii) Magnetic current I,.

---c Mom h e current

Fig. 4.26
Currents (I,) and (I,) add in arm A and nullify each other in arm
B a s in a r m B, the current equals difference of (I,) a n d (I,). For
I, =I,, current in loop A equals 21, or 21, and zero in loop B.
Magnitude of I, can be controlled by raising or lowering the loop D.
The capacitance between loop D and main line M is increased by
decreasing distance between D and M. This can be achieved by :
(a)changing insertion of loop inside the main line
( b ) changing length of the loop
Magnitude of I,,, can be controlled by :
( a ) changing position of the dipole
( b ) changing insertion of loop
The coupling mechanisms do not necessarily have to be holes in a
common wall. They can just as well be other transmission paths. A type
of strip-line directional coupler depends on electromagnetic coupling.
The inner conductors of two striplines run parallel to each other for a
distance of a quarter wavelength between common ground planes. The
spacing between the parallel conductors determines the amount of cou-
pling and the match depends upon their impedance. Excellent direc-
tivity is achieved a t the design frequency where the coupling region is
an exact quarter wavelength. The directivity is quite good over an oc-
tave bandwidth centred around the design frequency.
Reflectometer
Reflectometer is a device that can be used to make reflection meas-
urements. It makes use of the capability of a directional coupler to
separate incident power from the reflected power. Fig. 4.27 shows a
directional coupler configured as a reflectometer. Depending upon the
130 MICROWAVES AND RADAR

magnitude of the soupling factor, a frarlion of incident power is

1he remaining power travels
diverted to the reflectionless terminatio~~.
towards the device under test. The pcwsr reflected from the DUT then
can be measured a t port-3 of the coupler.

Fig. 4.27

4.6. Isolators and Circulators

Isolators and Circulators are used a t microwave frequencies to pro-
vide isolation. I t is important in majority of the applications a t
microwave frequencies that the coupling is a unidirectional affair.
This, for instance, applies to most of the microwave generators.
Isolators
Ferrites are used a s the main material in isolators, and the under-
lying principle on which the ferrite isolator works is the Faraday's
Principle or Faraday's Experimental Effect.

c(bJ l?zZi& @
Clrcular ly I
Ferrite
/
/
Tilted
polar~zed i / polarized
wave wave
Fig. 4.28
Faraday's principle states that "If a circularly polarised wave
(TEI1in a cylindrical waveguide) is made to pass through a ferrite rod,
whiEh has been influenced by an axial magnetic field B, the axis of
polarisation gets tilted in clockwise direction and the amount of tilt
depends upon the strength of magnetic field and geometry of the
ferrite". This principle is illustrated in Fig. 4.28.
A ferrite is a non-metallic material which has magnetic properties
similar to those of ferrous metals but a resistivity high enough to make
it an insulator. Commonly used ferrites include manganese ferrites
(MnFe203)and zinc ferrites (ZnFe203). Since these materials are in-
MICROWAVE COMPONENTS 131

sulators, electromagnetic waves can propagate in them and because

ferrites have strong magnetic properties, external magnetic fields can
be applied to them with Faraday's effect coming into picture, thus
making them useful for isolators and circulators.
Faraday's Rotation Isolator :The isolator consists of a piece of cir-
cular waveguide supporting the dominant TEl1mode with transitions to a
standard rectangular guide supporting the TElo mode at both ends, the
output end being twisted through 45".A thin pencil shaped ferrite is located
inside the circular guide, supported by polyfoam and the waveguide is sur-
rounded by a permanent magnet which generates magnetic field in the
ferrite core. To minimise reflections, the ferrite pencil is gradually tapered
at both ends, as shown in Fig. 4.29.

Output end
oriented by 4 5 0

Waveguide
Fig.4.29
Because of the applied DC magnetic field, a wave passing through
ferrite in the forward direction will have its plane of polarisation
shifted clockwise (45" in case of practical isolators) by the time it
reaches output end, a s shown in Fig. 4.30 (a). This wave then passes
through suitably rotated output transition end, that is, the rectangular
waveguide a t the right is physically oriented in such a way that the
plane of polarisation of incoming waves from the left coincides with
that of the usual TElo mode in this guide. The wave emerges with an
-
insertion loss of 0.5 1.0 dB .
A wave which tries to propagate through the isolator in the reverse
direction is also rotated clockwise, because direction of Faraday rotation
depends only on direction of DC magnetic field. Thus when the wave emer-
ges fi.om left, it cannot propagate through rectangular guide because of
dimensions of waveguide. Thus, under ideal conditions, no propagation
from right to the left, is possible as shown in Fig. 4.30 (b).Practically, the
attenuation is of the order of 30 dB. This reverse attenuation of an isolator
is called its isolation. SWR of such isolator will be less than 1.4, and a
 MICROWAVES AND RADAR
Forward wave

Tilt 45'

Input
Fig. 4.30(a)

Fig.4.30( b )
bandwidth of the order of 5 to 30% of center frequency. This type of
isolator is limited in its peak power handling capability to about 2 KW
because of non-linearities taking place in ferrite, resulting in phase
shift deviating from the ideal 45'. Fig. 4.31 shows the photograph of
some typical ferrite isolators.
Applications : Isolators can be used to improve the frequency
stability of microwave generators, such as klystrons and magnetrons,
where the reflection from the load affects the frequency. In such cases,
MICRO LVE COMPONENTS

Fig.4.31
the isolator is placed between microwave generator and the load so
that the energy is transmitted from the generator to the load with a
very small attenuation. On the other hand, energy of the reflected
waves resulting from load mismatch is highly absorbed by the isolator.
This prevents frequency instability of the oscillator.
Circulator
Another important application of ferrite is found in a microwave cir-
culator. A microwave circulator is
a multiport junction where the
power may flow from 1to 2, 2 to
3 a n d so on i n t h e direction

Although there is no restric-

tion on t h e number of ports, Transrn~tter Receiver
microwave circulators with four 0 o
ports are most commonly used. 1 3
A 4-port Faraday's rotational
circulator is shown in Fig. 4.33 and
is similar to Faraday isolator. The
power entering port-1 i n TE,,
mode is converted into TEll mode Antenna 2
because of gradual taper. This
power passes port-3 unaffected, is Fig. 4.32
rotated by 45' due to the ferrite, passes port-4 unaffected and finally
emerges from port-2, as it did in an isolator. Power fed from port-2
undergoes the same fate a s in an isolator but a final output comes out
134 MICROWAVES AND RADAR

from port-3 since port 3 is suitably aligned. Similarly port-3 is coupled

only tc? port-4 and port-4 t~port 1.This type of circulator is limited by
power handling capabilities, but is suitable as a low power device.
However, its use is restricted mostly to the highest frequencies, in the
millimeter range and above. Its characteristics are similar to those of
an isolator.
Circular I.

Taper Taper

---------.
- -,-'

Fig. 4.33
Wye Circulator :A Y or Wye cir-
culator is a recent development and
is shown schematically in Fig. 4.34.
A 3-pert version is shown 4-port cir- b~asedferrite
culators of this type are also com-
mon, consisting of tivo Ys connected
to each other and, therefore, having
four external ports.
If three waveguides or coaxia! Fig. 4.34
lines are arranged as shown in Fig.
4.34 and are 120' apart; a phase shift of 60" will ensure that the signal
is rotated to be coup!ed only to the next port and not the third port, i.e.
input from port-1 will be coupled only to port-2 and not to port-3 ;input
from port-2 will be coupled only to port-3 and so on. Thus circulator
properties are achieved.
A typical %port circulator (Wye type) handles only small powers,
has an isolation of the order of 20 dB or more, insertion loss of 0.5 dB
and an SWR of 1.25 in X band. A 4-port Y circulator has a n isolation of
more than 40 dE and an insertion loss of about 0.9 dB.
4.7. Waveguide Flanges
Waveguide flanges are used to couple sections of waveguides and
guide components. These flanges a r e designed to have not only
mechanical strength but also desirable electrical characteristics. The
flanges must have negligible power loss due to leakage and should not
produce any reflections throughout the frequency range. Flat butted
flanges are frequency-insensitive since the waveguides form a con-
tinuous passage a s shown i n Fig. 4.35. The guide ends must be
smoothened to avoid reflections and mechanical alignment should be
perfect to prevent leakage of power.
MICROWAVE COMPONENTS 136

Fig. 4.35
Choke flange :A choke flange shown in Fig. 4.36 is an improve-
ment over the plain flange described above.

Here a hg low impedance line is inserted in series with the waveguide

at the joint. Since the series line is short circuited at the far end, its input

point where-there
is a current node is
a point where the
actual contact be-
tween the flanges
occurs (at the mid-
dle of hg/2 length
line). Leakage and
power loss are thus
minimised. The
.flange's frequency L
sensitivity is mini- Fig. 4.37
mised by using low
characteristic impedance in hg/2 length line, so that its input im-
pedance does not rise rapidly in the vicinity of the design frequency. An
optimum design ensures a reasonable BW (say about 10 per cent of the
136 MICROWAVES AND RADAR

centre frequency over which SWR does not exceed 1.05). Fig. 4.37
shows a waveguide section terminated in a flange.
4.8. Rotating Joints
Rotating joints are frequently used in laboratory waveguide sys-
tems and in practical systems such as a radar, where a waveguide is
connected to a horn antenna feeding a paraboloid reflector which must
rotate for tracking.
The most important considerations in the design of such components
are low reflection, negligible power leakage and mechanical strength. Two
or more short circuited half wavelength lines are incorporated into the joint
in such a manner that the actual point of sliding contact is at or near a
current minimum. This is clearly illustrated in Fig. 4.38. In the waveguide
rotary joint, two sections of a
rectangular waveguide
operated in the TElo mode
are joined through a section
of circular waveguide
operated in the TMol mode,
in which fields a r e inde-
pendent of angular position
about the axis. To compen-
sate for the reflections occur-
ring due to junctions formed Fig.4.38
by circular and rectangular
waveguides, some kind of a tuner, say an inductive window, is used.
4.9. Attenuators
Microwave attenuators are used in every type of equiprhnt involv-
ing transmission, control or measurement of microwave energy. The
principal characteristics of concern to a microwave system designer
while choosing the right type for a given application include Range of
attenuation, Flatness with frequency, Average and peak power han-
dling capability and Temperature characteristics.
There are two broad categories of attenuators namely Fixed At-
tenuators that provide fixed attenuation and Variuble Attenuators offering
a variable attenuation. Fixed attenuators are commonly used in two types
of applications. One is in a calibration channel to establish a known signal
level. Flatness over required frequency range is important here. In the
second type, the device is used for impedance matching or as a buffer to
prevent interaction between two devices. Here, low VSWR is an important
requirement. Fig. 4.39 shows photographs of some typical fixed type at-
tenuators.
There are different types of variable attenuators. Variable at-
tenuation can be in the form of discrete steps or as a continuous varia-
tion. One of the commonly used types is the Rotary Vane Attenuator.
MICROWAVE COMPONENTS 137

Fig. 4.39. Coct.

 MICROWAVES AND RADAR

The rotary vans attenuator is a direct reading precision attenuator. A

functional diagram indicating the operating principle of this type of
attenuator is shown in Fig. 4.40 ( a )and (b)

(b)
Fig. 4.40
Basically, it consists of three sections of waveguide in tandem as
shown in Fig. 4.41 (a).A rectangular to circular wave guide transition
containing a horizontal attenuator strip is connected to a rotatable cir-
cular waveguide containing an attenuator strip and this in turn is con-
nected to a circular to rectangular wave guide transition containing a
horizontal attenuator strip. The attenuator strip for the rotary vane
attenuator is shown in Fig. 4.41 (b).
The incoming TElo mode is transformed into TEll mode in the cir-
cular waveguide by the rectangular to circular wave guide transition
with negligible reflections. The polarisation of the TEll mode is such
that the electric field is perpendicular to the thin resistive card in the
transition section. As such, this resistive card has a negligible effect on
the TEll mode. Since the resistive card in the center section can be
rotated, its orientation relative to the electric field of the incoming
TEll mode can be varied so that the amount by which this mode is
attenuated is adjustable.
MICROWAVE COMPONENTS 139

Res~stive
/ cards 7 \

Rectangular t b \ ~ o t a t i n ~section
circular waveguide of circular waveguide
transition

,-Aquadag coating-

Fig. 4.41
A with reference to Fig 4.40 (t),when all strips are aligned, the
electric field of the applied w,ive is normal to the strips and hence no
current flows in the attenuation strips and therefore no attenuation
occurs. In a position where the central attenuation strip is rotated by
angle 9, the electric field of the applied wave can be resolved into two
orthogonally polarised modes; one perpendicular and the other parallel
to the resistive card. That portion which is parallel to the resistive slab
will be absorbed, whereas the portion which is polarised perpendicular
to the slab will be transmitted. It can be proved mathematically, by
considering the analytic expression for the TEllmode electric field,
that the attenuation produced is given in dB by
a = - 20 log (sin2 9) = - 40 log (sin 9).

4.10. Phase Shifters

This is an instrument that produces an adjustable change in the
phase angle of the wave transmitted through it. Ideally, it should be
perfectly matched to the input and output lines ; and should produce
zero attenuation. There are a variety of designs for phase shifters. Two
types, the rotary phase shifter and a dielectric phase shifter are briefly
described here.
Rotary Phase Shifter
It is similar in construction to a rotary attenuator except that the
central resistive card is replaced by a half wave plate and the two outer
resistive cards are replaced by quarter wave plates. The quarter wave
plate converts linearly polarised TEll mode into a circularly polarised
mode and vice versa.
A slab of dielectric material is used for the construction of a quarter
wave plate. The proeagation constant PI, for the case, when the wave
140 MICROWAVES AND RADAR

is polarised parallel to the plate is greater than the propagation con-

stant pz for the case when the TEll wave is polarised perpendicular to
the plate, and also the length of the quarter wave plate 1 is so adjusted
that we get a differential phase change (PI - P2) I = 90'. The ends of the
dielectric slab are tapered to reduce reflections to a negligible value.
The half wave plate is similar in construction except that its length is
increased to produce a differential phase change of 180".
Rotatina section

A14 Plate A12 plate A14 plate

(a)

Dielectric.
slab
t
Y

Fig. 4.42 (a)shows the rotary phase shifter. The quarter wave
plates are oriented at an angle of 45" relative to the broad wall of the
rectangular wave guide. The rotation of half wave plate through an
angle changes the phase of the transmitted wave by an amount equal
to 29. This simple dependence of the phase change on a mechanical
rotation is the chief advantage of the rotary phase shifter.
Dielectric Phase Shifter
The principle of dielectric phase shifter is that the difference in
~ h a s eshift between two ~ o i n t sis determined by the velocity of
propagation and therefore is a function of the medium. Thus insertion
of a dielectric into the wave guide shifts the phase of the wave
propagating through it.

Fig. 4.43
MICROWAVE COMPONENTS 141

A simple variable phase-shifter may be constructed in a rectan-

gular wave guide by inserting a dielectric slab of thickness t , height h
and dielectric constant E, in such a manner that the h dimension is
parallel to the electric field ( T E l o mode) a s shown in Fig. 4.43.
~ ~

For a slab located a t x = x l , the propagation constant for the com-

posite structure is a function of E,, t , h and x l . The difference in phase
constant is found to be
h.t 2~x1
(P - Po) = 2x (E, - 1 ) - sin -.
a.b hg a
where Ago = Guide wavelength of the empty guide and & = Wavelength
of the composite structure. The differential phase shiR is given by :

Thus from above equation, it is clear that the slab has maximum
effect a t xl = a / 2 and no effect a t X I = 0.The two ends of the slab are
tapered to minimise reflections. The length of the tapers being ap-
proximately Ag/2. The slab is supported by means of two rods, which
are coupled with a screw. Thus rotating the screw produces lateral dis-
placement of the dielectric slab and hence phase-change.

4.11. Cavity Resonators

If a section of wave guide is closed a t its two ends by metal plates,
and the cavity so formed is then excited by a small probe or loop, it will
be found to have resonant properties similar to those of a section of
short-circuited transmission line. Resonators of this type, commonly
called cavity resonators, find extensive use a s reasonant circuits a t
microwave high frequencies. Their behaviour is analogous to that of
coil and capacitor combinations. But for microwave frequencies, cavity
resonators have the advantages of reasonable dimensions, simplicity,
high Q and very high shunt impedance. Cavity resonators are used, in
conjunction with waveguides, a s tuned circuits a t similar frequencies.
Their operation follows directly from the behaviour of waveguides.
Principle of Operation
Waveguides are considered from the point of view of standing
waves between lateral walls and traveling wave in the longitudinal
direction. If conducting end walls are placed such that the distance be-
tween the end walls equals nhg/2, oscillators will take place a t that
frequency, assuming that the resonator is suitably excited, as depicted
in Fig. 4.44.
As shown in Fig. 4.44, the presence of the short circuit a t the end
of the waveguide provides a standing wave pattern along the length of
the waveguide. The positions (i) and (ii) indicate the possible locations
 MICROWAVES AND RADAR
Suitable position for,
second short
circuitina wall /\
Short
circuiting
end wall

(ii) (1)

of the second short circuiting wall so that the pattern due to the first
wall is left undisturbed. Thus if the second wall is A/2 away from the
first, oscillations between these two walls will take place and continue
until all energy fed in is dissipated or the oscillations are sustained if
energy is constantly supplied as in a n ordinary tuned circuit. The
above stated conditions are satisfied by the space enclosed by conduct-
ing walls only a t a single frequency, or in other words, any such
enclosed space must have a resonant frequency.
Modes of Cavities
As in waveguides, it is possible for many different types of field
configurations or modes to exist in a cavity. To each mode, there cor-
responds a lesonant frequency that 1s determined by the particular
field configuration involved and the cavity dimensions. Each cavity
resonator infact possesses a n infin~tenumber of resonant frequencies.
The "dominant mode" is that field configuration having the lowest
resonant frequency, while the iemaining resonant frequencies are
referred to as higher order modes. The cavity modes can, in many
cases, be associated with waveguide modes.
Modes in a cavity a r e classified a s transverse electric (TE)or
transverse magnetic (TM) modes, corresponding as far as possible to the
analogous waveguide modes. The particular mode of any such class is then
commonly designated by three subscripts. Thus the resonator of Fig. 4.44
can be said to resonate in the TElol mode with the last subsrript indicating
that there is one half sine variation along the length of the resonator and
the first and second subscripts have the same meaning as in the case of
waveguides. However, such designation breaks down unless the cavity has
a simple shape, and for such complicated structures it is preferable to talk
in terms of resonant frequency rather than the mode. As shown in Fig. 4.44,
if the short circuit is placed at position (ii)instead of (i), then such a mode
is designated as TE 102 mode.
A cavity resonator possesses many more modes than does the cor-
responding waveguide. For example in the rectangular prism, there
are an infinite number of TELon modes for each of the three axes of the
MICROWAVE COMPONENTS

i
(a) Sphere

Electric A
fie
Magnetic
field i
(d) ~ o q h d u type
t
Electric field Magnetic field

!
El I

I Approximate distr~butlon
(b) Cylinder of magnetlc field I

I
Electric fleld
( e ) Reentrant type

Fig.4.45
 MICROWAVES AND RADAR

prism. Thus a triple infinity of modes exist in the rectangular prism,

corresponding to the single infinity of TElo waveguide modes. There-
fore, a t frequencies sufficiently larger than the frequency correspond-
ing to the dominant mode, it is found that the resonant frequencies of
cavities will be extremely closely spaced. This results in an impossible
situation if one wishes to obtain pure mode o~eration.At the same
time, it is an advantage if one desires to make it as easy as possible for
the cavity to resonate with an arbitrary exciting frequency.
Types of Cavity Resonators
Any enclosed surface, irrespective of how irregular its outline is,
forms a cavity resonator. Hence cavity resonator can take many forms.
The simplest one is a length of circular or rectangular waveguide short
circuited a t each end to form a cylinder or rectangular prism respec-
tively. One common defect with cylindrical, spherical or rectangular
prism cavity resonators is that their various resonant frequencies are
harmonically related. This defect is a serious drawback specially in a
the situation where the cavity is fed with a pulse of energy. The cavity
is supposed to maintain sinusoidal oscillations, but since the pulse of
energy is rich in harmonics and that the cavity is able to oscillate at
harmonic frequencies, the output of the cavity is still in the form of
pulses. This explains the reason why most practical cavities are odd
shaped. These odd shaped cavities ensure th2t the various oscillating
frequencies are not harmonically related, and therefore the harmonics
are attenuated. Refer to Fig. 4.45
In a spherical cavity the opposite sides are brought close together to
form a re-entrant structure. Such spherical structures are especially used
in klystron tubes. In such an ar-
rangement, the electric field
is very strong in t h e gap
forwed by the re-entrant sec-
tions, thus permitting effec-
tive interaction with electrons
passing across this gap. They
a r e known a s re-entrant
resonators because one of the
walls re-enters the resonator
shape.
Cavity resonators can
also be derived from co-axial
lines. Such a resonator could
I/ Jb

c 2
Fig. 4.46
7' -z

be obtained by placing short

circuits a t the two ends of a length of co-axial line. The length is a
multiple of a half wave length.
Resonant Frequency of a Rectangular Cavity
Consider a rectangular cavity of height (b),width (a)and length ( c )
as shown in Fig. 4.46. The solution for the rectangular cavity may be
obtained directly from the corresponding waveguide solutions. For the
TE or TM mode, the propagation constant is given by
MICROWAVE COMPONENTS 145
I
where KO= 2nfo/u,. We require pmn e = pn where p is an integer in
order for the cavity to be a multiple of a half guide wavelength long.
Thus when d is specified, Pmn is given by

However, this relation is true only for discrete values of KO.Only if

KO= K, where Kmnpis given by

will the above equations be satisfied. These particular values ofKOgive

the resonant frequencies of the cavity; i.e.,

where u, is the velocity of light

The expression for resonant frequency of a rectangular cavity

resonator for TM modes is same as the one given for TE modes.
In case of circular cavity resonators, the expressions for the
resonant ti-equency in case of TE and TM modes are given as under :

where u, = Free space velocity of electromagnetic waves

a = Internal radius of guide
d = Length of guide
(ha1)= Eigen value of Bessel function

Also,
]:""iK(I
f =A -- + (7) 1 ( f o r TMnmpmodes)

The expressions for semi-circular cavity resonator having same

radius as that of a circular cavity resonator are same as those given
above for circular cavity resonators.
146 MICROWAVES AND RADAR

Q of Cavity Resonators
The 'Q'of a cavity resonator may be e?pressed by tbe fundamental
expression :
Energy s h r e d
Q = 211
Energy lost in each cycle
Roughly speaking, energy is stored in the volume of the resonator
and dissipated through its surface ; hence it follows that the shape
giving the highest volume-to-surface area ratio is likely to have the
highest Q. When high Q is the primary requirement, we use spherical,
cylindrical or rectangular cavity resonators. A well designed cavity,
with its inner walls gold or silver plated, would have an unloaded Q in
the range of about 2000 for re-entrant cavity to 100,000 for a spherical
one and a value exceeding 40,000 for the spherical cavity when it is
loaded.
The energy stored is proportional to the square of the magnetic flux
density, integrated throughout the volume of the resonator, while the
energy lost per cycle in the walls is proportional to the skin depth and
to the square of the magnetic flux density integrated over the surface
of the cavity.
When. a cavity is tuned by means of a screw or a sliding piston, its
Q will suffer, and this should be taken into account. The Q decreases
because of the extra area due to the presence of the tuning elements, in
which current can flow. However such tuned cavity resonators find
wide application in the microwave range.
Another method for tuning a cavity is by generally introducing a
dielectric material. The frequency of resonance would vary depending
upon the depth of insertion. This introduction will have an effect of
changing the resonant frequency, since the signal wavelength in the
resonator is affected. The wavelength will be reduced inside dielectric
because the velocity of light inside dielectric is less than in air, and so
will the size of the cavity required a t any given frequency. However,
the disadvantage of dielectric tuning is that the dielectric materials
have significant losses a t microwave frequencies and therefore the Q of
the cavity will be reduced by their introduction.
Applications
The purpose of cavity resonators is the same as tuned LC circuits
or resonant transmission lines, the only difference being that cavity
resonators are used a t much higher frequencies, since they have the
same overall frequency coverage as waveguides. They may be input or
output tuned circuits of amplifiers, tuned circuits of oscillators, or
resonant circuits used for filtering or used in conjunction with mixers.
Ope of the many applications of the cavity resonators is as a cavity
wakemeter, used as a microwave frequency-measuring device.
MICROWAVE COMPONENTS 147

4.12. Wavemeters
Cavities have many uses in microwave circuits - both as circuit
elements and as measuring instruments. Fixed cavities are used as
elements in filters and for frequency control in microwave oscillators.
Tunable cavities are used a s wavemeters and receiver pre-selectors.
A cavity is made tunable by varying one of its dimensions, usually
the length, and calibrating the variation. Tunable resonators or
cavities may be used either as :
(i) Transmission type wavemeter
or (ii) Reaction type wavemeter
Transmission type wavemeter
As is evident from the equivalent circuit of Fig. 4.47 ( b ) ,maximum
energy reaches the load only when the cavity is tuned to its resonant
Generator . Cav~ty I Load

(a) Wave meter

Fig. 4.47
frequency. The transmission wavemeter is coupled via a probe or an
E-plane tee to the main guide or line, and has an output coupling loop
with a detector. As stated above, the indicator connected to the detector
probe will show maximum output a t resonant frequency, this frequen-
cy of oscillation may be read off from the wavemeter.
Absorption-type or Reaction-type wavemeter
This type ofwavemeter is as shown in Fig. 4.48 and is the one more
frequently used.
T h e distance d is
made either zero or ap-
proximately a guide half
wavelength so that an ef-
fective short presents it- d Wavegu~de
self a t t h e waveguide
wall whenever the cavity Generator -+ b -~oad
is far from resonance.
The total effective cavity Fig. 4.28 ( a )
impedance is effectively
148 MICROWAVES AND RADAR

in series w i t h t h e
load and hence a dip
in the load power oc-
curs as the
wavemeter is tuned
through resonance.
The m a g n i t u d e of Generator
the dip in power as
- -Load

the cavity is tuned Fig. 4.48 ( b )

through resonance
depends upon the magnitude of cavity impedance and consequently on
the quality factor Q and the distance d.
Cavity wavemeter
A typical cavity wavemeter is shown in fig. 4.49. The axis of cavity
is made perpendicular to the main wave guide, and the coupling from
the cavity to the waeeguide is achieved through a smaller circular hole
or iris. A block of polyron (absorbing material) on the back of plunger
prevents false resonance in the back cavity by damping out any oscil-
lations that might exist. It should be noticed that a s the resonant
cavity is made smaller, the back chamber gets larger and may intro-
duce false resonance. By proper design of cavity, the instrument may
be made to indicate wavelength directly on a micrometer head attached
to a plunger. Calibration of this wavemeter is done by determining the
resonant frequency for different micrometer settings. A calibration
chart is drawn from these readings and the wavemeter calibrated ac-
cordingly.
Iris, rFlange

Fig. 4.49
The accuracy obtained using wavemeters is sufficient for most
practical purposes-the errors being somewhere between 1part in lo2
and 5 parts in lo5. Skill of operator, accuracy of initial calibration of
wavemeter, temperature stability and its Q are some of the factors
which may improve the accuracy of the wavemeter. In particular, the
cavity should have a very high Q of the order of 1,000 to 50,000. In
general, a higher Q will yield greater accuracy.
MICROWAVE COMPONENTS 149

However, a high Q wavemeter is not always the most desirable ;

since accuracy of the order of 1in lo3 is quite adequate and a high Q
resonator is most tiresome to use unless frequency is known to a qer-
tain extent. The ultimate limit of accuracy depends upon initial
calibration of wavemdter.
4.13. Hybrid ring (Rat-race)
It is an arrangement performing similar functions as magic tee but
is constructionally different from magic tee.
The arrangement has a piece of rectangular waveguide bent in E-
plane to form a complete loop whose median circumference is 1.5 hp. It
has four orifices with separations as shown in Fig. 4.50 ; from each of
which a wave guide emerges as shown.
If there are no reflections from
terminations of the arms, any arms is
coupled to two other arms and not to
the remaining arm.
If a signal is applied to arm 1, it
will divide equally-one half travell-
ing in clockwise direction and other
half travelling i n anti-clockwise
direction. Signal reaching 4 covers
same distance, clockwise as well as
anti-clockwise and hence addition
takes place a t that arm resulting in
some signal travelling out of arm 4. L
Similarly a t 2, the portion of sig- Fig. 4.50
nal travelling clockwise travels A,/4,
and that travelling counter-clockwise covers 5 $/4 and the two por-
tions are again in phase resulting in addition a t arm 2 and propagation
down the arm 2 also takes place.
As far as arm 3 is concerned, the signal reaching it covers hp in
counterclockwise direction and hp/2 in clockwise direction ;hence the
two components reaching 3 are in phase opposition and they cancel out
resulting in no output from arm 3.
Similarly we can show that arm 3 is connected to arm 4 and 2 but
isolated from arm 1, and that the behaviour is somewhat similar to
that of magic tee.
Hybrid ring and magic tee may be used interchangeably with
magic tee having the advantage of smaller size but suffering from the
disadvantage of requiring internal matching which is not needed by
rat-race if thickness of ring is correctly chosen. Further, hybrid ring
seems to be preferable a t short wavelengths since its dimensions are
less critical.
150 MICROWAVES AND RADAR

4.14. Slotted Line

Slotted line remains the foremost device for precise measurement
of VSWR, Insertion loss and Impeda~ce.These are available in both
coaxial as well as waveguide versions. The slotted line is a piece of a
coaxial line or section of a waveguide with a long slot. The slot is in the
outar conductor in case of coaxial line and in the centre of the broad
wal! in case of the waveguide slottgd line. In all measurements with the
slotted lilAe,the line is used for measurement of standing wave pat-
twns t.f the electric field intensity as a function of the longitudinal posi-
tior in the guiding structure. Tine slotted line system (Fig. 4.51) has a
movalls carriage extending into a longitudinal slot in a section of the
coaxial line or waveguide a s the case may be. A sampling probe
n~nuntedon the movable carriage picks up the signal, amplifies it and
d.sp:ays the same in terms of the cquare law response of the barretter
or crq atal detector. The detector-carriage assembly is also known as the
tra relling detector.

Inner
1
conductor
p Utelectr~c
support
-Slot
tL Corr~age
1

Fig. 4.51
The sources of measurement error include probe tuning, har-
monics and other spurious signals, signal frequency modulation and
residual VSWR of the slotted line itself. The residual VSWR of the
slotted line is considered as a single discontinuity caused by reflections
from all discontinuities in the slotted line system.
One of the most common measurements done with the slotted line
is the measurement of VSWR. Fig. 4.52 shows the typical test set up
MICROWAVE COMPONENTS 151

w
Detector

Slotted lme
Meter

Matched
Termlnation

RF
Source +"Iso1ator
+
m '
Detector

I
Slotted hne
Meter

I DUT

(b)
Fig.4.52
used to measure VSWR with a slotted line. To carry out such a meas-
urement, the slotted line is first terminated in a matched load as shown
in Fig. 4.53 (a)and is applied a 1kHz modulated signal from the RF
source. The probe depth is set as recommended by the manufacturer.
In case the probe is tunable, which is usually the case, it is set for
maximum signal output. The modulation frequency is also fine tuned
to get maximum signal output. This happens when the modulation fre-
quency is tuned to the tuned response of the SWR meter. Next, the
matched termination is replaced by the device under test (DUT) as
shown in F'ig. 4.52 (b).The probe is moved along the slotted line to get
maximum'SWR meter indication again. The SWR meter range switch
and Gain control are then adjusted to get a reading of SWR = 1on the
SWR meter. The probe is moved again to get the minimum SWR read-
ing. This minimum reading is then the VSWR for the device under test.
If it is not possible to get the minimum reading, the next higher meter
range is selected and the last step repeated. This would usually become
necessary only if the DUT has a relatively higher value of VSWR.
4.15. Striplines and Microstrips
Both Striplines and Microstrips are miniature transmission lines
which were developed mainly to take full advantage of miniaturisation
brought about by semiconductor microwave devices and microwave in-
tegrated circuits. Infact, striplines and microstrips are so packaged as
to offer easy interconnection to semiconductor microwave devices.
Stripline
Stripline has evolved from a co-
axial cable. It can be thought of as a
compressed or flattened co-axial cable
(Fig. 4.53) whose edges have been cut
away. A stripline consists of a pair of
flat metallic-ground planes sepatated Fig. 4.53
152 MICROWAVES AND RADAR

by a dielectric, in the middle ofwhich is embedded a thin metallic strip.

It is possible to use more than one independent strips with a common
set of ground planes and the dielectric. The dielectric used is often
teflon, alumina or silicon.
Like co-axial lines and waveguides, striplines can also be used to
build other microwave components such a s Isolators, Circulators,
Duplexers etc.
The guide wavelength (Ag) in case of a strip line can be computed
from
Lo
kg =
where (k)is the dielectric constant of the dielectric used between the
conducting planes.
Microstrip
Microstrip has evolved f?om a parallel wire line. It consists of a top thin
strip sitting on a dielectric that rests on a single ground plane as shown in
Fig. 4.54. The thin strip and its M~crostrip
image below the ground
plane provide analogy to the D~electrtc
open wire line. Here also, it
is possible to use more than
one strips on a common Ground plane
dielectric and ground plane.
Typically, dielectric thick- Fig.4.54
ness may vary from 0.1 to 1.5
mm while the metallic strip may be as thin as'l0 pm.
The guide wavelength (Ag) in a microstrip can be computed from

kl + hz
where k e f = -
2
k l = Dielectric constant of dielectric used
k2 = Dielectric constant of medium surrounding
the microstrip
As compared to a stripline, microstrip has the advantage of having
a relatively simpler construction and easier integration with semicon-
ductor microwave devices. On the other hand, microstrip has a greater
tendency to radiate from irregularities and sharp corners with the
result t h a t t h e isolation between microwave circuits ioined bv.
MICROWAVE COMPONENTS 153

Microstrips and Striplines vs. Waveguides

In addition to their small size and easier integration with semicon-
ductor microwave devices, microstrips and striplines have an edge over
waveguides and co-axial lines for low and medium powers upto 100
GHz. Above 100 GHz, they are not practicable due to significant in-
crease in costs and losses. As compared to waveguides, these have the
advantage of a greater bandwidth. Waveguides can be operated over a
frequency range of 1.5 : 1. The bandwidth is limited in case of
waveguides by cut-off wavelength a t the lower end and the frequency
at which higher modes may propagate at the upper end. There is no
such restriction in microstrips and striplines. Bandwidths greater than
2 : 1 are common. However, when compared to waveguides, these have
lower Q , higher losses and lower power handling capability. Circuit
isolation though good in striplines, is no match to that of waveguides.
4.16. SAW Devices
SAW devices make use of propagation of surface acoustic waves
(SAW) on the surface of a solid piezoelectric material. The commonly
used piezoelectric materials are Quartz and Lithium niobate. To make
a SAW resonator, thin lines are etched on a metallic surface that has
been electrodeposited on a piezoelectric substrate. Etching is done
using photolithography or electron beam techniques. SAW devices are
seldom used below 50 MHz. The upper frequency limit dapends upon
the photoetching accuracy. If the acoustic wave velocity is taken to be
3000 mls, the required finger separation a t 10 GHz would be 0.3 pm.
This is currently the upper frequency limit. It may be mentioned here
that in a SAW resonator, an applied RE' signal gives rise to travelling
waves in both directions. The resulting standing wave is maintained
only a t the frequency a t which the distance between the adjoining
fingers is equal to a n integral multiple of an acoustic wavelength along
the surface of the material. Fig. 4.55 shows a simplified sketch of a
Acoustic Fmgers
wavelength
t
rwr \\ 1

Piezoelectric
substrate

Fig. 4.55
154 MICROWAVES AND RADAR

typical interdigitated SAW resonator. When such a device is used as a

filter, only those frequencies close to the resonant frequency of the
SAW resonator will be allowed to pass through which makes SAW
device a narrow band bandpass filter. To build a n oscillator using SAW
resonator, one only needs to place it in the feedback path between the
input and output of a n amplifier. The arrangement is shown in Fig.
4.56.

Fig. 4.56

SOLVED PROBLEMS

Problem 4.1 :A 10 mW signal is applied to a 20 dB Directional

Coupler. Determine the power available at the coupled port.
Solution : Input Power = 10 mW
Coupling Factor = 20 dB
If (Pi)and (PC)are respectively the input and coupled powers,

E)
Then Coupling Factor (in dB) = - 10 log - = 10 log -
I:;[
Now,
(i:)
10 log - = 20

. .

Pi-- 100 which

-
gives PC= -Pi = -
10 m W = 0.1 mW
PC 100 100
Therefore, Coupled Power = 0.1mW = 100 pW

Problem 4.2 : Compute the power available a t the straight-

through-port output (output port of the primary line) i f the Direction
coupler has an Insertion Loss of 0.4 dB and the input power is 10 mW.
MICROWAVE COMPONENTS 155

Solution : I f (Pi)and (Po) are respectively the input and output

powers,
Then Tnsertion Loss = 10 log (Pi/Po)

That is, I0 log ($1- 0.4

which gives

or --
lo - Antilog 0.04= 1.096
Po
which gives

Problem 4.3 : I n a certain Direction Coupler set-up, the power

measured at the coupled port is 20 d B below the input power whereas
the power measured at the isolated pcrt is 50 d B below the input power.
What should be the Directivity ~f this cogpler ?Also compute the power
at the isolated port if the power measured at the coupled port is 100
micro-watts.
Solution :Directivity is given by :
Directivity (in dB) = Isolation (in dB) - Coupling Factor (in dB)
=50-20=30dB
Also, from the definition of Directivity,

where
Directivity = 10 log -
(2)
PC= Power at the coupled port
Piso = Power at the isolated port

J:[
10 log - = 30

Now, PC= 100 w,therefore, Piso = -'' - - loo- 100 n~

1000 - 1000 -

,Problem 4.4 : The input power to a 20 d B Directional Coupler is

10 dBm. If the Directivity of the coupler is 30 dB, determine the power
available at the output port of the primary line.
Solution : Coupling Factor = 20 dB
Directivity = 30 dB
156 MICROWAVES AND RADAR

Input power = 10 dBm

If (Pi)is input power in milliwatts, then
10 log Pi = 10
or Pi= 10 mW
Now, Isolation (in dB) = Directivity (in dB) + Coupling Factor (in dB)
=20+30=50dB
Power at the coupled port
= l o - 2 0 = - 10dBm=0.1mW=100pW
Power a t the isolated port

Therefore, power available at the output port is given by,

10 - (0.1 + 0.0001) milliwatts
= 10 - (0.1001) = 9.8999 mW

Problem 4.5 :A directional coupler has a coupling factor of 10 dB.

An input signal of 5 mW is applied. Determine the Directivity of the
Directional Coupler if the power measured a t the isolated port is 10
PW.
Solution :Input power = 5 mW
Coupling Factor = 10 dB

Therefore,

where Pi= Input power

(2)
10 log - = 10

PC= Coupled power

Pi
This gives = 10, PC= 0.5 mW
, ,
Power measured a t isolated port, Piso= 10 pW

Directivity
go)
= 10 log - = 10 log -

= 10 log 50 = 17 dB
(Yo(')

- - -

Problem 4.6 :Compute the lowest resonant frequency of a rectan-

gular cavity resonator having following dimensions :
Width, a = 2 cm ;Height, b = 1cm ;Length, d = 3 cm.
Solution :The cavity resonator has the lowest resonant frequency
for the dominant mode which for a rectangular cavity is (TElol)mode.

The resonant frequency = -

2
MICROWAVE COMPONENTS 157

For TElol mode, m = 1,n = 0, p = I

Therefore, resonant frequency,

Problem 4.7 :T h e T E l l l mode resonant frequency i n a certain rec-

tangular cavity resonator i s 10 GHz. Determine the corresponding
resonant frequency for T M l l l mode.
Solution :The resonant frequency for a T M mode in a rectangular
cavity resonator for a given integral values of m, n and p is the same as
that of a T E mode for same values of m, n and p.
Therefore, T M l l l mode resonant frequency = 10 GHz.

Problem 4.8 : The least resonant frequency for a certain rectan-

gular cavity resonator is found out to be 6 GHz. If the broad and narrow
dimensions of the cavity resonator are 4 and 2 c m respectively, deter-
mine its length.
Solutim : Resonant frequency is least for TElol mode.

Now,

or

or

or

or

or

Therefore, length of cavity resonator = 3.2 cm

Problem 4.9 :For the rectangular cavity resonator of Problem 4.8,

verify that the length o f resonator is half of the guide wavelength.
158 MICROWAVES AND RADAR

Solution :For the data given in Problem 4.8, the cut-off wavelength
can be computed from :

The operating wavelength,

The Guide Wavelength, hg can now be computed from :

Ag = 6.4cm
or Ag/2 = 3.2 cm = length of cavity resonator

Problem 4.10 :A circular wave.pide cavity resonator having an

internal diameter of 8 cm resonates a: 10 GHz in TMoll mode. Deter-
mine the length of resonator given that (haol)= 2.405.
Solution :Internal diameter = 8 cxn
Internc! radius = 4 cm
Operating frequency, fo = 10 GHz

where (ha)is the Eigen value of the Bessel function, ( a )is internal radius
and ( d )is the length.
Substituting the given values,
MICROWAVE COMPONENTS 159

Problem 4.11 : Compute the resonant frequency of a circular

resonator ~f internal radius 6 cm and length 5 cm for fa) TMOl2mode
fb) TElll modegiven that (haol)= 2 . 4 0 5 and (hallt)= 1.841.
Solution :Internal radius, a = 6 cm
Length, d=5cm

Problem 4.12 : I f the cavity resonator ofproblem 4.11 is a semi-cir-

cular one instead of a circular one with the same internal radius, what
would be the resonant frequencies in the two cases mentioned in Prob-
lem 4.11 ?
Solution :The expressions for determining the resonant frequency in
case of TE and TM modes for a semi-circular resonator are the same as
those given earlier for a circular cavity resonator. If the internal radius is
the same, the resonant frequency of a given TE or TM mode will be same
for both circular and semi-circular cavity resonators.
160 MICROWAVES AND RADAR

Review Questions
1. Briefly describe the methods of probe and loop coupling for
exciting waveguides.
2. What are the different types of flange couplings used for
joining waveguide sections ? Which is preferred where ?
3. What is the basic purpose of using a circulator ? With the
help of a typical application configuration, briefly describe its
principle of operation.
4. Differentiate between a n E-plane tee and a n H-plane tee.
Briefly describe the use of a magic tee as an isolator.
5. Name the commonly used techniques for impedance match-
ing in waveguides. Briefly describe the characteristics of each
of them.
6. Define the Coupling Factor and Directivity of a directional
coupler. With the help of a sketch, briefly describe how a
waveguide directional coupler can be used as a Reflectometer.
7. Briefly describe the functioning of a cavity resonatsr. What
are their major applications ?
8. Write short notes on the following :
(a) Ferrite phase shifter
( b ) SAW devices
(c) Strip lines and Microstrips
( d ) Microwave attenuators
9. What is a slotted line used for ? Briefly describe the steps to
be followed for measuring VSWR of a given device using a
slotted line.

Problems
1. A 10 mW signal is applied to a Direction Coupler having a
coupling factor of 30 dB. Compute the power available a t the
coupled port. If the power a t the isolated port is 10 nW,
compute Isolation and Directivity. (10 pW, 60 dB, 30 dB)
2. The input power to a 25 dB Direction coupler is 20 dBm. If the
Directivity of the coupler is 40 dB, determine the power available
at the output port of the primary line. (99.684 mW)
3. The TMlZ1mode resonant frequency in a certain rectangular
cavity resonator is observed to be 9 GHz. Determine the
resonant frequency for TEIz1 mode. (9 GHz)
4. A semi-circular waveguide cavity resonator having an internal
radius of 4 cm resonates at 10 GHz in TMoll mode. Determine
the length of the resonator given that (haol) = 2.405.
(1.565 cm)
MICROWAVE COMPONENTS

SELF EVALUATION EXERCISE

Multichoice Questions
1. A coupling probe is placed a t a point
( a ) where the electrical field has maximum intensity
( b ) where the magnetic field has maximum intensity
( c ) where the electric field has the least value
(dl none of these
2. A coupling loop is placed a t a point of
( a ) maximum magnetic field intensity
( b ) minimum magnetic field intensity
( c ) maximum electrical field intensity
( d ) minimum electrical field intensity
3. Mark the correct statement
( a ) Capacitive windows are readily adjustable
( b ) Windows insertion should be much larger than the
guide wavelength
( c ) A tuning screw more than ( h / 4 ) long causes capacitive
susceptance
( d ) A tuning screw less than ( h / 4 ) long causes capacitive
susceptance
4. A better way of achieving variable susceptance in a waveguide
is by using
( a )inductive windows ( b )capacitive windows
( c )resonant diaphragms (dl tuning screws
5. The windows shown in the waveguide of Fig. 4.57 (a)present
( a )an inductive susceptance ( b )capacitive susceptance
( c ) a resonant circuit (dl ntpe of these
6 . Fig. 4.57 ( b ) shows the cross-sectional view of a waveguide.
It is
( a )a capacitive window ( b )an inductive window
( c )a resonant circuit ( d )an inductive post
162 MICROWAVES AND RADAR

7. Microwave components used to rclnect branch waveguide to

the main waveguide or transm~ssioniine are knovn as
( a )waveguide tees ib) directional couplers
( c )slug tuners ( d )posts
8. Branch waveguide is connected i r series with the main
waveguide by
( a )an E-plane tee ( b )an H-plane tee
( c )a magic tea ( d )none of those
9. In a shunt tee, the axis of the side arm is in
( a ) the plane of E-field of the main waveguide
( b ) the plane of H-field of the main waveguide
( c ) the plane normal to the H-field of the main waveguide
(dl the plane normal to the E-field and H-field of the main
waveguide
10. Magic tee is nothing but
( a ) a modification of E-plane tee
( b ) a modification of H-plane tee
(c) a combination of E-plane and H-plane tee
(dl two E-plane tees connected in parallel
11. Refer to the microwave component shown in Fig. 4.58 (a).I t
is
( a )an E-plane tee ( b )an H-plane tee
(c) a magic tee (dl a duplexer
12. Refer to the microwave component shown in Fig. 4.58 (b).It is
( a )an E-plane tee ( b )an H-plane tee
( c ) a circulator ( d )a magic tee

13. In a waveguide tee, the signals entering the two side ports
are in phase and have power levels of P1 and P2. If it is an
H-plane tee, then the power level of the signal a t the exit port
will be
( a )PlP2 ( b )P1 - P2
( c )P1 + P2 (dl P 1 / P 2
MICROWAVE COMPONENTS 163

14. If in Q.13, the waveguide tee is an E-plane tee, the signal

power level a t the exit port will be
(a)PI+ Pz ( b )PI- P2
(c)PIP2 (dlP 1 / P 2
15. Refer to the waveguide tee shown in Fig. 4.59. If the signal
power level entering port-C is 10 watts, the power level exit
from port-D will be
(a)0 ( b )10 watts
(c) 5 watts (d)indeterminate from given data

Fig. 4.44

16. The ratio of power coupled to the secondary arm in the

preferred direction to the power in the main line in a 20 dB
directional coupler will be
( a )20 (b)0.05
(c)0.01 (d)0.1
17. One of the following microwave components can possibly be
used for transmission and reception of microwave signals
using the same antenna.
(a)Isolator ( b )Directional coupler
(c)Slotted line (d)Circulator
18. One of the following microwave components is used with
microwave bench for making measurements such as standing
wave ratio (SWR), guide wavele~gthetc.
(a)Slotted line ( b )Magic tee
(c) Circulator ( d )Directional coupler
19. One of the following is a passive microwave device.
(a)IMPATT diode ( b )W N N diode
(c)Klystron ( d )Stripline
20. A microstcrip is analogous to a
(a)co-axial line ( b )parallel wire line
( c )rectangular waveguide (d)circular waveguide
21. A stripline is analogous to a
( a )parallel wire line ( b )co-axial line
( c )circular waveguide ( d )rectangular waveguide
22. A SAW device is not used a s one of the following :
( a )Delay line ( b )Filter
( c )Oscillator ( d )Amplifier
23. The acoustic wave velocity on a certain SAW device is 3000
m/s. If this device has a resonant frequency of 4 GHz, its
finger separation would be
( a )750 nm ( b )12 pm
( c )750 pm ( d )indeterminate from given data
24. Surface acoustic waves can propagate in only one of the
following materials/devices.
( a )Microstrip ( b )Gallium arsenide
( c )Indium phosphide ( d )Lithium niobate
25. MMIC stands for
( a ) Monolithic Microwave Integrated Circuits
( b ) Micro Machined Inegrated Circuits
( c ) Miniature Microwave Integrated Circuits
( d ) Microwave Monolithic Integrated Circuits.
 Microwave Tubes
assive nlicrowave conlponents were discussed in the last chap-

P ter and active microwave devices, which include microwave

tubes and semiconductor microwave devices, form the subject
matter of the present chapter and the one to follow after this.
Microwave tubes can be classified into three broad groups. The first
group contains gridded tubes such as triodes that we are all familiar
with. The second group contains those tubes where there is a brief
interaction, though sometimes more than once, between the electron
beam and the RF field. Klystrons belong to this category of microwave
tubes. In the third group are the tubes in which there is a continuous
interaction between the electron beam and the RF field. This includes
Magnetrons where the continuous interaction is ensured by a magnetic
field, Travelling Wave Tubes (TWT)where the same is ensured by an
electric field and other similar ones. While klystron is an amplifier, its
oscillator counterpart is Reflex klystron. The oscillator counterpart of
a TWTAis the Backward Wave Oscillator (BWO). Cross Field Amplifier
(CFA) conlplements Magnetron oscillator. This chapter begins with a
brief discussion of factors that limit the performance of gridded tubes
a t ultrahigh frequencies and beyond, to be followed by a detailed
account of each one of the tubes listed above.
The topics covered in this chapter include :
5.1. Limitation of 5.7. Traveling Wave
ridded tubes at Tubes
f igh frequencies
5.2. Klystrons
5.8. Backward Wave
Oscillator
5.3. Multi-cavity 5.9, Comparison of
klystrons Microwave tubes
5.4. Reflex klystrons 5.10. Gyrotron
5.5. Magnetron
5.6. CrossFieldAmpli-
fier
166 MICROWAVES AND RADAR

5.1. Limitations Of Gridded Tubes

at.High Frequencies
A gridded tube, invariably a triode, when used at microwave
frequencies suffers from two serioils problen~s.It may be mentioned
here that the problem of noise is more severe at higher frequencies and
triode is the least noisy of all tubes. That is why, a gridded tube is
invariohly a triode when it is meant for use at microwave frequencies.
Going LscE to the problem areas, one of the problems originates from
fii~itevalues of interelectrode capac~tances(grid-cathode, grid-anode
and anode-cathade)andleadinductances. The other is dile to the transit
timr of slectrons to travel from cathode to anode becoming a significant
proporticn of the time period of the signal being handled. The two
problems are briefly described below :
Interelectrode Capacitances and Lead Inductances
Fig. 5.1 shows the equivalent circuit of a triode tube at microwave
fre,yt.cies. The interelectrode capacitances and lead inductacces are
of the order of 1 to 2 pF and 15 to 20 nH respectively. The shunt
Impedances due to interelectrode capacitances and series impedances
due to lead inductances become significantly low and high respectively
to render these tubes unusable at microwave frequencies. Refinements
have been done in the d ~ s i g nand fabrication of these tubes with the
result that these tubes, such as Disk seal tube. are stilt in use upto the
lower end of n~iaowak-espectrum.

Fig. 5.1
Transit Time Effects
When the transit time becomes an appreciable part of the time
period of the RF signal, it starts playing havoc with the signal. The
electrons leaving cathode when RF input is at its maximum negative
value experience a different veloci*;y profile than those leaving the
cathode when the RF signal is at its maximum positive value (or the
grid is least negative). Due to this velocity modulation, whenever there
are more electrons on the cathode side of the grid than those on the
anode side of the grid, electrons must flow out of the grid or current
MICROWAVE TUBES 167

must flow into the grid terminal. The reverse occurs i . e . the electrons
flow into the grid or current flows out of grid ternlinal when there are
less electrons in the grid cathode space than those in the grid anode
space. The RF drive source gets loaded in both cases with the loading
being proportional to square of frequency. As the frequency increases,
the transit time effect becomes more and more severe. In fact, there is
no solution to transit time effect in gridded tubes. The solution to this
problem was found in a new set of microwave tubes that made use of
the transit time and the consequent velocity modulation in their opera-
tional principle.
5.2. Klystron
A Klystron, for its operation, depends upon velocity n~odulation
which leads to density n~odulationof electrons. One of the earlier forms
of velocity modulation devices is the Two cavity klystron amplifier,
represented by the schematic of Fig. 5.2. It is seen that a high velocity
electron beam is formed, focused and sent down along a glass tube to a
collector electrode, which is at a high positive potential with respect to
the cathode. As it is clear from the schematic of Fig. 5.2, a two cavity
klystron amplifier consists of a cathode, focusing electrodes, two
buncher grids separated by a very small distance forming a gap A, two
catcher grids with a small gap B followed by a collector. The significance
of buncher and catcher grids will be clear from the discussion in
following paragraphs.
The input and output are taken from the tube via resonant cavities
called Buncher and Catcher cavities. The region between buncher grids
and catcher grids is called drift space. The electron beam passes gap A
in the buncher cavity to which RF signal to be amplified is applied and
is then allowed to drift freely without any influence from RF field until
it reaches gap B in the output or catcher cavity.
Buncher Catcher

output
Electron
bunches\

electrodes

'I'l11
Fig. 5.2
The first grid (focusing grid) controls the number of electrons in
the electron beam and serves to focus the beam. The velocity of electrons
168 MICROWAVES AND RADAR
in the beam is determined by the beam accelerating potential. On
leaving the region of first grid, the electrons pass through the grids of
buncher cavity. The grids of the cavity allow the electrons to pass
through, but confine the magnetic fields within the cavity. The space
between the grids is referred to as interaction space. When electrons
travel through this space, they are subjected to RF potential at a
frequency determinated by the cavity resonant frequency which is
nothing but the input frequency. The amplitude of this RF potential
between the grids is determinedby the amplitude ofthe incoming signal
in case of an amplifier, or by the amplitude of feedback signal from the
second cavity if used as an oscillator. Klystron tube used as an oscillator
is discussed in the latter part of this chapter. If all goes well, oscillations
will be excited in the second cavity which are of a power much higher
than that in the buncher cavity.
As it is seen that the cavities are of re-entrant type and are also
tunable. Further more, they may be integral or demountable; in the
latter case the wire grid meshes may be connected to the rings external
to the glass envelope, and cavities may be attached to the rings.
Although the drifi space is quite long and the transit time in it is put to
use, the gaps must to be short so that the voltage across them does not
change significantly during the passage of a particular bunch of
electrons. Having a high collector voltage helps in this regard.
It has already been said that when electrons pass through gap A ,
they are influenced by the RF voltage across this gap. However, the
extent of this effect on any electron will depend on the voltage across
the gap at the time the electron passes this gap. It thus becomes
necessary to investigate the effect of the gap voltage gap on electrons
passing through the individual electrons passing through the gap.
Velocity Modulation
Consider a situation when there is no voltage across the gap.
Electrons passing it are unaffected and continue on to the collector with
the same constant velocities they had before approaching the gap.
Sometime later, after an input has been fed to the buncher grid, an
electron will pass gapA at the time when the voltage across this gap is
zero and going positive, let this be the reference electron y. This
reference electron is unaffected by the gap, due to the fact that it has
the same slope on theApplegute diagram of Fig. 5.3. as electrons passing
the gap before any signal was applied to the buncher cavity.
Consider another electron z passing the gap slightly later than y
as shown. In the absence of gap voltage, both electrons would have
continued past the gap with unchanged velocity and, therefore, neither
would have caught up with the other. In presence of positive voltage
across gap A , however electron z is accelerated slightly, and given
enough time, will catch up with the reference electrony easily before
they approach,gapB. similarly, consider another electronx passing gap
A slightly before the reference electron, and is retarded by the negative
voltage present at that instant across the gap. Since electron y was not
MICROWAVE TUBES 169
so retarded, it has an excellent chance of catching up with electron x,
before gap B, and this is done as shown in Fig. 5.3.

- Bunches

Bunching
limits

Fig. 5.3
As electrons pass the buncher gap, they are velocity modulated by
the RF voltage existing across this gap ; such velocity modulation is not
sufficient in itself for amplification by the klystron. However, as ex-
plained with reference to the Applegate diagram, the electrons are given
an opportunity to bunch in the drift space. When an electron catches
up with another one, it may simply pass it and forge ahead. On the other
hand, it may exchange energy with the slower electron, giving it some
excess energy, and the two bunch together and move on with the average
velocity of the beam. This phenomenon is very vital to the operation of
klystron tube as an amplifier. As the beam progresses further down
along the drift space, the bunching becomes more complete, as more
and more of the faster electrons catch up with bunches ahead. Even-
tually, the current passes the catcher gap with quite pronounced
bunches and, therefore, varies cyclically with time. This variation in
current density (often called current modulation) enables the klystron
to have a significant gain.
It is evident from the applegate diagram that bunching can occur
only once per cycle, centering on the reference electron. The limits of
bunching are also shown. Any electrons ariiving slightly after the
second limit are not accelerated sufficiently to catch the reference
electron and that the reference electron cannot catch up with any
electron passing through the gapA just before the first limit. Bunches,
therefore, arrive at the catcher grid, once per cycle and then deliver
energy to this cavity. The catcher cavity is excited into oscillations at
its resonant frequency, which is same as the input frequency and alarge
sinusoidal output can be obtained because of the flywheel effect of the
170 MICROWAVES AND RADAR
output resonator. The parameters that need to be considered or op-
timised for operating klystron amplifier with maximum efficiency are :
1. Drift space
2. Signal amplitude
3. D.C. anode voltage
Above three factors when properly adjusted give maximum effi-
ciency.
Mathematical Analysis of Two Cavity Klystron
The following assumptions are made in the mathematical treat-
ment to follow :
1. Electron transit angle in buncher and catcher grids is very
small.
2. Vl c Vo, where Vl is the amplitude of RF signal voltage at
buncher grid, Vo is accelerating voltage.
3. Space charge effects are negligible.
4. Electron beam density is uniform throughout the length, i.e. no
loss of electrons takes place in buncher and catcher grids.

input output
Fig. 5.4
g. 5.4, d = Interaction gap
s = Drift space
Vo = Accelerating voltage

-
We know from the very well known relation that

Voltage at grid 1= Vo
Voltage at grid 2 = Vo + VI sin o t i
MICROWAVE TUBES 171

In the above relation, vo represents eIectron velocity at grid 1 of

buncher grid and Vl sin at1 is the feedback voltage to the buncher cavity
from catcher cavity.
1
Also -rnv12 = e(V0 + Vl sin dl) (5.2)
2
(Electrons come out of buncher grid 2 at time ti and velocity v I)
From equations (5.1) and (5.2),we get
v i r vodl+!Vi/Vo)sinoti (5.31
Eq. (5.3) is obtained provided V1<<Vo. This is an equation of
velocity modulation for centre of bunch electron, i.e. when atl= 0,
Substituting, at1 = 0 in equation (5.3), we get
v1= vo
n
Velocity maxima and minima respectively occur at at = -
2

n
For at1 = +-, ~1 mar = vo 41 + (VI/VO)
2

Let T be the transit time inside the drift space and t2 be the time
a t which electron reaches catcher grid,
Then, tz = ti + T
From Eq. ( 5 . Q we get
wt2= oti + wT
where at2 = Transit angle at catcher grid
wtl = Transit angle when electrons leave
bmcher grid
S S
T=.--=
vi vo d l + (Vi/Vo) sin cot
From Eq. (5.6) and (5.7)

WS
where -= cc is called transit angle for centre of bunch electron
VO
and at2 - otl = transit angle inside the drift space
172 MICROWAVES AND RADAR
Power output a t catcher grid :Let V2 sin at2 be voltage at catcher
grids, hence energy given by an electron is equal to
- eV2 sin at2 = W '.
(Negative sign indicates that energy is given by the electron)
Average energy given by an electron to the field (per electron)
2%

P -I
- 1
au-27T
2"
0
2 at21 d (at11
(- e ~ sin

From Eq. (5.9) and using binorninal equation, we get

Pa" = - eVz Jl (x) sin a (5.12)

where

x is a bunching parameter for the klystron, Ji (x)= Bessel function of

the first order.
Fig. 5.5. gives a graph between J1 ( x ) and x.
Let there be N electrons reaching the catcher.
Then, energy transferred to the catcherlsec
= N.Pav = Pat

= - Ne V2 Ji (x) sin a [since current = - Ne =lo]

Fig. 5.5
PaC= - IoV2Jl (x) sin a
D.C. Power Pdc = ZoVo
Pac -
Efficiency q = - Va Jl(x) sin a
---
pdc VO
MICROWAVE TUBES 173
For maximum efficiency,
J1 ( x ) = 0.58 at x = 1.84 (Refer to Fig. 5.5)
and sin a = 1

Therefore, a = 2xn - n/2. ['.' V2 can never exceed Vo]

--vz (0.58) = 58%.

- [For Vz = Vol
Vo
Value of accelerating voltage for maximum output and
hence maximum efficiency :

to be maximum, x = 1.84 and

Substituting for (x) and (a), we get

mu8
From Eq. (5.1) Vo =

Now, ):(
vo=

which gives, Vo = 0.284 x lo-''

For n = 1, Vo is maximum

Therefore, (Vo),, = 0.284 x lo-"

(37J
From Eq. (5.16)
174 MICROWAVES AND RADAR

53.Multi-cavity Klystron
It is a common practice to one or more additional cavities between
the catcher and buncher cavities. Refer to three cavity klystron tube of
Fig. 5.6. Here oscillators are excited in the middle cavity by the partially
bunched electron stream passing the gap B; in this way a voltage is
produced across B that also acts on the electron stream. By detuning
the additional cavity so that its gap cffers an impedance having an
inductive component (i.e. resonant frequency slightly above the signal
frequency), the phase of the voltage across B is related to the electron
stream at B in such a manner as to canse further velocity modulation.
This considerably increases the voltage amplification of the tube and
likewise raises the efficiency. It is d s o possible to increase the
bandwidth of a klystron amplifier b; employing one or more inter-
mediate cavities that are properly :uned.
A four-cavity klystron is the commonly used klystron amplifier
tube. Fig. 5.7 shows the cross-sectionaldrawing of afour-cavity klystron
(Type VA-849). Fig. 5.8. shows thz photograph of the same.

Buncher Intermed~are
cavlty cavity

Fig. 5.6
Performance and Applications
A rnulticavity klystron can be used as a medium or high power
amplifier in tile U.H.F. and microwave ranges, for both continuous and
pulsed operation. These tvhes are available over the range 25(! MHz to
about 60 GHz, with typical maximum CW output powers of 100 KW in
the V.H.F. range and a maximum of about 250 KW in the Xband. Pulsed
peak powers u ~ t 25o MW are also available.
MICROWAVE TUBES 175

Fig. 5.7
176 MICROWAVES AND RADAR

The power gain of nwlticavity klystron varies from about 30 dB at

U.H.F. to over 60 dB in thexband, with bandwidths from 8 to 60 MHz

-. ... . .
TUNING

Fig. 5.8
and mechanical tuningranges from 300 to 600 MHz. The size ofklystron
depends upon the operating frequency which determines the cavity size,
length of drift space and also the magnet size, and lastly the output
power. Typically, a high power U.H.F. klystron may be upto 7 ft long
and may weigh over 250 kg. Multicavity klystron amplifiers are very
noisy in practice. This dictates their applications, which are obviously
in transmitters rather than receivers. Such klystrons are used as power
output tubes in U.H.F. television transmitters, in troposphere scatter
transmitters, in some satellite communications ground stations etc.
These are also used for pulsed power requirements.
The two cavity klystron oscillator is used as a power oscillator in
the frequency range from 5 to 50 GHz, with output powers rangingfkom
2 to 10 W ;but as much as 200 W CW is also available from commercial
devices. The main applications are in CW doppler radar, for pumping
parametric amplifiers and as frequency modulated oscillators in high
power microwave links. The biggest advantage of two cavity klystrons
is the relatively high CW power ofwhich they are capable of as compared
to their small size. The biggest disadvantage is that they require two
cavities and each has to be retuned separately if a frequency change is
required. The cavities are high Q cavities with narrow bandwidths, and
thus individual tuning is awkward. Therefore, two cavity klystrons are
generally used in fixed frequency applications. When frequency is to be
MICROWAVE TUBES 177

varied, reflex klystron, discussed in the next section, is used instead. It

has only one cavity resonator which makes tuning process a lot easier.
5.4. Reflex Klystron
A reflex klystron is a low power, low efficiency microwave oscil-
lator, illustrated schen~aticallyin Fig. 5.9. It consists of an electron gun
sinlilar to that of a multicavity klystron, a filament surrounded by a
cathode and a focusing electrode at the cathode potential. The suitably
formed electron beam is accelerated towards the cavity, which has a
high positive voltage applied to it and thus acts as an anode. After
passing the gap in the cavity, electrons travel towards repeller electrode
which is at a high negative potential.

Focussing Output
electrode 7

Cavity
(anode)

Fig. 5.9.
The electrons are repelled back from midway of the repeller space
by the repeller electrode towards the anode. If conditions are properly
adjusted, then the returning electrons give more energy to the gap than
they took from it on forward journey, thus leading to sustained oscilla-
tions.
Operational Principle
It is assumed here that oscillations can begin through noise or
switching transients. In order to understand the operation, i t is ad-
vantageous to consider specific electrons, passing the gap for the first
time at selected times. Here reference electron taken is y and it passes
the gap without being affected by the voltage across it and travels
towards the repeller, and is returned back to anode without touching
the repeller. The fast electrons come nearer to the repeller than slow
ones, and therefore take a longer time to return to the resonator gap
than electrons which did not approach the repeller so closely.
Fig 5.10 shows the Applegate diagram for a reflex klystron. It
shows the path of reference electron y, rather the position of y at any
instant in the repeller space. Its path of course is straight out and then
straight back along the same line. Consider an electron x which passes
the resonator gap (on its way out) slightly before the reference electron.
178 MICROWAVES AND RADAR
Had there been no gap voltage, x would have returned before the
reference electron. However, due to presence of RF voltage, the electron

4
I Bunching
I T cycles

limits

Fig. 5.10.
will be velocity modulated. Electron x, as seen is in fact accelerated due
to the positive field available to it and so comes closer to the repeller as
compared to reference electron y. As seen from Fig. 5.10, it is quite
possible that y will catch up with electron x as they enter back into
resonator gap. Similarly electron z leaving the gap slightly after the
reference electron y experiences a negative field. Therefore electron z
does not reach as close to repeller compared to electron y, before it is
returned back to resonator gap. There exists every possibility that
electron2 also catches up with electron y as it enters back into the gap.
Thus all the three electrons x, y and z return back nearly a t the same
time into the resonator gap. This ofcourse, needs optimisation of various
tube parameters such as Transit time, Repeller electrode potential and
SO on.

The system is analogous to a 2-cavity klystron where velccity

modulation is converted into current modulation after the electrons
have left the gap on their outward journey. The bunching in reflex
klystron is not as complete as in a multicavity klystron, because there
are quite a few electrons arriving at the gap out of phase and contribut-
ing to the high noise and low efficiency of the device. However, bunching
is sufficient between the bunching limits, to make operation possible.
Again, one bunch is formed per cycle of oscillations around the reference
electron, and that these bunches deliver energy to the gap.
Transit Time
For sustained oscillations the time taken by the electron to travel
into the repeller space and back to the gap, called transit time, must
have an optimum value. This factor is not so important in a klystron
amplifier, but it assumes a great importance in case of a reflex klystron.
MICROWAVE TUBES 179
The most optimum departure time is obviously centred on the
reference electron which is at 180' point of sine wave voltage across the
resonator gap. Actually no energy goes into velocity modulation of the
electron beam. It takes some energy to accelerate electrons, but just as
much energy is gained from retarding electrons. Since there are as
many retarded electrons as those which are accelerated, the total
energy outlay is nil. From the above discussion, it is evident that the
best possible time for electrons to return to the gap is the one at which
the voltage then existing across the gap will apply maximum retarda-
tion to them. This is when the gap voltage is positive maximum. This
causes electrons to fall through the maximum negative voltage between
the gap grids, thus giving up the maximum amount of energy to the
gap. Thus it appears that, the best time for electrons to return to the
gap is at the 90"oint of the sine wave gap voltage. Referring to Fig.
5.10 returning of electrons after 1; cycles satisfies the above require-
ments. When generalised, the transit time in repeller space is given by

where n = any integer

Mathematical Analysis for Reflex Klystron
A

Repeller

-1
Fig.5.11
Refer to Fig. 5.11,

Voltage difference between pointsA and B is

VR- (Vo + VI sin at) = VR - Vo
Electric field between points A and B is
180 MICROWAVES AND RADAR

(VR- V O )
Force = - eE = e -
s (5.21)
md2x
Also, Force = Mass x Acceleration = - (5.22)
dt2
Wherexis any distancebetween pointsA andB startingfrompoint
A.
Then from Eqs. (5.21) and (5.22)

At t =tl,v =Vl.
Integrating expression (5.23), twice, we get
x=
4%- Vo) ( t - tl)2 + v l ( t - t l )
2ms

e ( V . - YO)
or vl (t2 - t i ) = - (t2 - t l ) 2
2ms
or v1=- e (VR - V O )(t2 - t l )
2ms
- 2m svl
or (t2 - t l ) =
e (VR- V O )
- 2msviw
or o(t2 - t l ) =
2 (VR- Vo)
or 0t2 = 0t1 -
2wmsvo
~ ( V R Vo)
-
(I
+%sin
2Vo
ot)

where a' corresponds to centreif bunch electron and

Here bunching parameter is

and for sustained oscillations

MICROWAVE TUBES 181

Relationship between accelerating voltage and repeller

voltage :

mvo2
Now, uo = 4 T w h i c h gives Vo = -
2e

Also

mvoL
This gives, Vo = -
2e

Bunching parameter, Vl a'

x' = -
2Vo

i t implies

and
Therefore, efficiency

q = -Pm

m
= W Jl(x1)

pdc
Maximum efficiency is obtained at n = 1
182 MICROWAVES AND RADAR
Relation between repeller voltage and frequency of opera-
tion of reflex klystron :
From Eq. (5.30),

Differentiating Eq. 5.32 with respect to a,

Substituting the value of (VR- Vo)from Eq. (5.32),

From Eq. (5.33) it is found that frequency and hence the transit
time is dependent on the repeller and anode voltages, so that both must
be carefully adjusted. Once the cavity has been tuned to the correct
frequency, both the anode andrepeller voltages are adjusted to give the
correct value of transit line from the data supplied by the manufacturer.
A significant rise in the power output shows that correct values of
voltages have been reached. Every time the cavity is retuned, the anode
and repeller voltages have also to be adjusted accordingly.
MICROWAVE TUBES 183

Modes : Fig. 5.12 shows the range of anode and repeller voltage
conlbinations for which oscillations will take place. The shaded areas
indicate such suitable conlbinations. It is viewed that oscillations are
possible for a range of these voltages, and yield transit times between
:1 and :5 cycles. Each value of n is said to correspond to a different
mode for the reflex klystron oscillator. The mode used mostly is the most
convenient one ;the lower (corresponding to lower value of n) the mode,
the larger the output power, which is obviously an advantage. The
voltages required for lower mode are higher leading to insulation
problems and the possibility of lower efficiencies. As a result, the most
commonly used are those corresponding to n = 2 or n = 3

Anode voltage -
Fig. 5.12

c5 I Negative repeller voltage -

Fig. 5.13
 184 MICROWAVES AND RADAR
By suitable choice of parameters, we can have any one of the
available modes of oscillation. Fig. 5.13 shows oscillation frequency with
repeller voltage as a variah!e parameter
The values taken in Fig. 5.13 are just to explain the concepts. In
Fig. 5.13, i t is seen that maximum power output is achieved when
n = 2. For any mode, maximum power is obtained when the repeller
voltage (&longwith anode voltage being taken as constant) gives the
optirnumvalue of the transit time. When the repeller vcltage is on either
side of the qorrect value, the output power is less than the maximum.
Fig. 5.13. dcpicts another important characteristics of a reflex klystron.
Ifwt- assume the cavity to be exactly.r,esonant at 10 GHz, then the period
of each q c l e is = 10-"' sec With mode-2 in operation, the
(10 x l o 9 )
comhintlhion of anode and repeller voltages must, therefore, provide a
trailsit time of 2: x lo-'' = 275 ps. If the i-epeller voltage is a little low,
the c.ctv;l transit time would be shghtly larger than 275ps and the tube
would operate at a different frequency assuming that the cavity
bandwidth allows this. Thus reducing the repeller voltage alone reduces
the frequency of oscillation of the reflex klystron and vice versa.
Performance :Reflex klystron is one of the most commonly used
microwave devices. These are commercially available for frequencies
ranging from under 4 to over 200 GHz. Typical power outputs range
from 10 mW to 2.5 W, maximum available figures ranging from 2 W a t
5 GHz to 1.5 W at 12 GHz. Efficiencies are fairly low (under 10 per cent)
but this is not so significant for low power applications.
T u n i n g :The method consists of moving in or out a small screw
which in turn operates on an arm thht tightens or loosens small bellow;
which tightens or loosen a wall of the tuning cavity thus changing the
size and frequency of resonance of the tuning cavity. Electronic tuning
is the most common type of tuning in which the repeller voltage is
varied.
Repeller protection : It is very important to ensure that the
repeller of the reflex ldystron never draws current ;if it does ever do so,
i t would be quickly destroyed due to the impact of high velocity electrons
as well as over heating. To avoid this, a cathode resistor is often used
to ensure that the repeller can never be more positive than the cathode,
even if the repeller voltage fails. Also the system is so arranged that
negative voltage is applied to the repeller before positive voltage
appears at the anode. Usually a protection diode is connected across the
klystron (with the diode's anode to the repeller, and its cathode to the
klystron cathode), so that again the repeller voltage cannot become
positive.
Applications :Reflex klystron finds extensive applications in the
following microwave systems :
1. Local oscillator in microwave receivers
2. Signal source in microwave generators
MICROWAVE TUBES 185

3. Frequency modulated oscillator in low power portable

nlicrowave links
4. Pump oscillator for parametric amplifiers
5.5. Magnetron
Magnetron is a type of high power microwave oscillator and it
forms the basis of many a microwave radar transmitter systems even
today. This microwave generator makes use of magnetic field for
producing oscillations at microwave frequencies. Microwave mag-
netrons now find many applications apart fromradar. They are capable
of producing megawatts of peak power in the centimeter wavelength
range and may be operated at wavelengths extending down to mil-
limeter range.
All magnetrons consist of some form of anode and cathode operated
in a magnetic field that is normal to the electric field between the
cathode and the anode. The magnetic field causes electrons emitted
from the cathode to move i~.,curved paths. In cake the flux density of
the magnetic field is sufficiently strong, the electrons do not strike the
anode but return to the cathode and the anode current is cut off.
There are three broad caiegories of magnetrons.
1. Negative resistance magnetrons
2. Cyclotron frequency magnetrons
3. Traveling wave magnetrons

Negative resistance magnetrons make use of a static negative

resistance present between two anode electrodes/plates. Cyclotron fre-
quency magnetrons depend on synchronism between an alternating
component of electric field and periodic oscillations of electrons in a
direction parallel to this electric field. The third type, i.e. traveling wave
magnetrons depend on interaction of moving electrons with a traveling
electro-magnetic field traveling with a constant angular or linear
velocity. Must modern magnetrons come under the category of traveling
wave magnetrons.
Constructional features of a cavity magnetron
The Cavity Magnetron or a Traveling Wave Magnetron or simply a
magnetron has cylindrical construction employing aradial electric field,
an axial magnetic field and an anode structure with permanent cavities.
Fig. 5.14 shows the cross-section of a Hole-And-Slot magnetron. Here
the cylindrical cathode is surrounded by the anode with cavities and
thus a radial D.C. electric field will exist. The magnetic field due to a
horse-shoe type of magnet used in conjunction with the magnetron
produces axial magnetic field, i.e. magnetic lines of force are at right
bngles to the plane of radial electric field. The magnetic field is also D.C.
and since it is perpendicular to the plane of the radial electric field, the
magnetron
- is referred to as a Crossed Field Device.
The output of the magnetron is taken from one of the cavities by
means of a co-axial line as is clear from Fig. 5.14 or through a waveguide
186 MICROWAVES AND RADAR
dependingon the power and frequency. It is to be noted that the output
coupling loop leads to a cavity resonator to which a waveguide is
connected and thus the whole output from this magnetron is taken v i a
a waveguide. The material used for the anode is normally copper
regardless of its shape. Fig. 5.15 shows the photograph of a typical
cavity magnetron complete with its permanent magnet.

output

YLLLLLLY

Fig. 5.14

PERMANENT

AIR COOLING

HEATER-CATHODE 1
VOLTAGE TERMINALS

Fig. 5.15
There are a number of resonant cavities in a magnetron and,
therefore, it must have an equal number of resonant frequencies and/
or modes of oscillation. The mode that is used must be self-consistent.
For example, it is not possible for an 8 cavity magnetron (the most
MICROWAVE TUBES 187

practical type) to employ a mode in which phase difference between

adjacent anodes is 30" ; this leads to a total phase difference of
30 x 8 = 240" ; meaning thereby that first anode pole piece is out of
phase from itself by 120". It can be seen that the smallest phase
difference between two adjacent pole pieces obtainable with an 8 cavity
magnetron is 45" or ~ / which 4 results in an overall phase shift of
360" or 2n radians which is self-consistentas is required. This 7c/4 mode
is seldom used in practice because i t does not give suitable charac-
teristics and the more commonly used mode is the rc-mode. Modes are
discussed in detail a little later.
From the viewpoint of construction, two types of structures are
used. These are :
1. Parallel plate magnetron
2. Cylindrical magnetron
We shall do mathematical analysis of each of the types.
Parallel plate magnetron
Fig. 5.16 shows a simple diagram used to calculate the value of the
cut off magnetic field strength required by an electron to just graze the
anode.

Fig. 5.16
Let Vo be the voltage applied between anode and the cathode of a
parallel plate magnetron. An electron leaving the cathode is under
influence of an electric field as well as a magnetic field which acts into
the plane of the paper and perpendicular to it.
Electric field strength between the plates,
188 MICROWAVES AND RADAR
Force on an electron in a magnetic field.
F = QvR sin 0 (5.35)
where 0 is the angle made by the magnetic field with the direction of
electric field.
If B=9O0,F=QvB (5.36)
and the path traced by an electron is a circle. In Eq. (5.35),
Q = Charge of electron
v = Velocity with which electron moves
B = Magnetic flux density
Now the force actingon the electron given by Eq. 5.35. is equivalent
to a centrifugal force exerted due to motion of electron in a circular path
of radius r and is given by :

m v m v
so that r = - -=- -
'Q'B e 'B
where Q = e = Electronic charge
and r = Radius of circle in which the electron moves
Angular velocity of electron is given by :

Vectorially, the force on electron is given as

-
f=(i;xB+E)
where v, B, E are velocity vector, magnetic field strength vector and
electric field vector respectively. Hence magnitude of the force f i
along Z-direction is
f i = m . a z = - e [E+vxB1 (5.40)
and force (magnitude) along X-direction is
fx = m.ax = + evzB (5.41)
In above equations ax and az are values of accelerahn in X and Z
directions respectively.
Eqs. (5.40) and (5.41) may be rqwritten as
dvz eB
-=- vz = oevz
dt m
where

and
MICROWAVE TUBES 189

Differentiating Eq. 5.42 yields

d2uZ
-=- dux
dt2 we dt
=we 2 uz (5.43)
Eq. (5.43) is a differential equation of second degree and its most
general solution may be written as
v, = cl sin wet + c2 cos a t (5.44)
Now at t=0,vz=Oqc2=0
Therefore, v z = cl sin a t (5.45)
Differentiating (5.45) w.r.t. t , we get
dvz
-- - C l w e cos wet
dt
Equating Eqs. (5.46)and (5.42)we obtain
eVo
C l @ cos aet = - wevx + -
m .d

So that the solution of (5.44) is obtained as

vz = -eVo sin wet
mdae
From Eq. (5.42), we get
eVo 1 dvZ- eVo
(5.49)
v x = oemd
--- we ' dt a md [I- cos wet]
Eq. (5.49)is obtained by differentiating Eq. (5.48)and rewriting it
in above equation

That is,

Since
1 dvz
Therefore, --= cos wet
we dt
Now Eq. (5.48) can be reproduced here as
- = - eVO sin met
v z = dz
dt mdo,
190 MICROWAVES AND RADAR
Integrating Eq. (5.48) with respect to (t)

This gives,

Therefore, z=- ev02 [I- cos met]

mdme
Similarly from Eq. (5.49),we have
dx
ux=-=- eVo (1- cos a t )
dt me md
Integrating, (5.49) w.r.t. (t) we get
eVo
x=-t-- eVo sin met + C'
memd mdmz
Andat t=O, x=O Cf=O

SO that
x =-
eVo
memd [t - - 1
sin met

Eq. (5.52) could be rewritten here as

Now for value of z to be maximum, we have

coS met = - 1 3 a t = 7t
2eVo - 2eVo/d
Therefore, zmax = --
mme2d

At d = zmax., the electron just grazes off the parallel plate so that
we have the expressim for cutoff value of magnetic field, Be as
Be=- (5.56)
Eq. (5.56) is an expression for cutoff magnetic field in case of
parallel plate magnetron.
Cylindrical magnetron
A cylindrical magnetron consists of two cylindrical electrodes -the
inner one being the negative cathode and the outer one being the anode.
Let the radius of Anode = b and the radius of cathode = a
The path of motion of the electron under influence of electrical and
magnetic fields is shown in Fig. 5.17. The magnetic field acts in a
direction perpendicular to the direction of electrie field which is radial.
MICROWAVE TUBES 191

Inner cathode
Direction of motlon of
electron emitted by cathode

Fig. 5.17
Torque = Time rate of change of angular momentum (5.57)
Force acting on electron is given as
-
F = - e (V x B ) (5.58)
+component of force, i.e. FQis given as
FQ= evpB 5.59)
and torque in +direction is given by
~ . F =Qep$ (5.60)
where p is any distance from center of cathode
Angular momentum can be computed as follows :
Angular momentum = Angular velocity x Moment of inertia
Now, Angular velocity = &t!
dt
Moment of inertia of an electron = m p 2
Therefore, Angular momentum = mp 2 x 4! (5.61)
dt
Now, Torque = Time rate of change of angular momenturn

Integrating Eq. (5.62) w.r.t. t

d+ = 0
p = a i.e. at cathode, angular velocity, -
dt
 MICROWAVES AND RADAR
"

d$ eB
Therefore, - --

Maximum value of angular velocity is

Now from Law of conservation c i energy, sum of potential and

kinetic energies at any point remains the same.
Potential energy of an electron = - eVo
and kinetic energy at any instant = $ .7tv2
2
Now -evo+'mv
2
=O
2
* -- - e V o =m
In3
z ( v P2 + v l 2 ) (5.67)
where vp and ve are velocity com?onents in directions p and $ respec-
tively.
42
= dt and v+= p dt 2

So that ev= 1rn [(s) (,) 2

+ p2 %!
2
]
When
eV='[(*)
2 dt
B = Bmt off,
2

+.'. me2k-$)] 2

(5.69)

velocity *o
dt
=

and
MICROWAVE TUBES

Fig. 5.18
e ~ ~ ~ b ~
If b > > a , then Vo=-
8m

Mechanism of Oscillations in a Magnetron

mode oscillations :For oscillations to sustain in a magnetron
~r:
oscillator, the phase difference ($) between two adjacent anode poles
should equal (2m/N)where ( N )is the number of cavities and (n)is any
integer. For an eight cavity magnetron and n = 4, ($) equals II: radians.
This gives rise to n-mode oscillations (Fig. 5.19).

Electric field Y

Fig.5.19
When RF field is absent, the path followed by electrons a and b is
shown by dotted lines, but this path is modified by the presence of RF
electric field. This RF field is present in the individual resonators as
well, but is omitted here for simplifying the discussion and also because
no significant contribution is made by it to the magnetron operation.
194 MICROWAVES AND RADAR
The important point here is that each cavity acts like a short circuited
h / 4 transmission line, with each gap corre:por,r.ing to an open circuit
or voltage maxima point in the resulting s t a n ~ i n gwave pa4tern. The
electric field, also extends into anode interaction space.
The n~echanisn~ by which oscillations ares mtained in a magnetron
can be understood by examining the behaviour 3f electrons enlittzd from
the cathode, when acted upon sii~~ultaneocsly by DC electric field and
axially acting magnetic field in the absence of radio frequency oscilla-
tions.
An electron emitted from the cathode under above conditions is
accelerated towards the anode by the radial electric field produced by
the DC anode voltage. In the absence of a magnetic field, the electron
is drawn towards the anode as shown by path a in Fig. 5.19. But as the
electron gains velocity, the axial magnetic field present exert.; a force
on it. When magnetic field is not very strong, the electron is deflected
as shown by b in Fig. (5.19) as has been explained before.
As we have already seen that in Fig, 5.19, the dotted paths a and
b show the path followed by the electrons in absence of RF alectric field.
The RF field associated with the oscillations acts on the electrons and
hence modifies the orbits followed by the electrons. Electron a is so
located with respect to these RF fields that its tangential velocity
experiences an opposition by the field. This electron, hence is slowed
down by the oscillations and hence it gives out energy to the oscillations.
Moreover, since electron a loses its velocity, the magnetic field exerts a
reduced deflecting force with result that electron moves toward the
anode, instead of being turned back toward the cathode, as shown by
solid line path a.
The relationship between DC anode voltage and magnetic field is
made such, that the tangential velocity of the electron causes the time
required by electron a to travel from position 1 to 2 is approximately
half cycle of the radio frequency oscillations. Then as the electron a
reaches point 2, the electric field reverse polarity from that shown in
the figure. The result is that electron a continues to be slowed down,
and goes on drifting towards the anode. However velocity acquired by
electron a does not change appreciably as it approaches the anode,
because the energy that is acquired by this electron fromfalling through
the DC anode-cathode voltage is largely delivered to the oscillations
instead being used to increase the velocity of electron. This electron
finally strikes the anode surface after having delivered to the oscilla-
tions a major part of the energy represented by its fall through the DC
cathode anode-potential.
Let us consider another electron b in Fig. 5.19, which is emitted
under such conditions that it is accelerated by the radio frequency field.
Instead of being slowed down, this electron gains energy and, therefore,
it deflects more sharply than in previous case. As a result the path
followed by this electron b is shown by the solid line in Fig. 5.19. This
electron is not a desirable one as it extracts energyfrom the oscillations.
But since it does not linger for a long time, it does not have much time
to absorb a significant amount of energy. However it bombards the
MICROWAVE TUBES 195
cathode on its return with a velocity corresponding to the energy gained
from oscillating field causing back heating of the cathode. This tends to
deteriorate the cathode and about 5% of anode power of an operating
magnetron is lost in this manner in heating the cathode.
There is also a focussing mechanism which helps to keep the
working electron in step with the fields in the interaction space. The
action is such that nlaximum possible energy is delivered by the
working electrons to the oscillations. As an example, let us consider an
electron c in Fig. (5.19)which delivers some energy to the oscillations,
but was emitted alittle late to be in correct position to provide maximum
contribution. This electronc is acted upon by a radial component of field
from the oscillations and also the tangential component. This radial
field is in such a direction that it aids the DC anode voltage, i.e. acts in
same direction as DC anode voltape. This results in an increase in
velocity of electron c and hence helis it in catching up with electron a
which is in optimum position. Similarly, an electron d, which has gone
beyond the optimum position electron a , is slowed down giving out
energy to the oscillations. This slowing down takes place due to its being
faced up with a radial field that opposes the anode voltage acting on it
and hence it is attracted less strongly towards the anode. As a result,
it falls in step with the optimum position. This focusing action is similar
to a velocity modulation that causes electrons such as c and d to form
a bunch centred about electron a. This is the bunching action of the
magnetron known as Phase Focusing Effect.

Anode
&Cathode

Fig. 5.20
The ultimate result of the various actions that take place is to cause
the electrons to be confined to spokes, as shown in Fig. 5.20, one for each
pair of anodes. In case of n-mode, these spokes rotate at an angular
velocity equivalent to 2 poles per cycle, and a definite fraction of
electrons emitted from the cathode travel out through these spokes
regularly delivering energy to the oscillations until they reach the anode
and are absorbed. Also simultaneously, electrons enlitted in the por-
tions of cathode between the spokes are however returned very quickly
to the cathode. Though these dectrons are harmful, as they absorb some
106 MICROWAVES AND RADAR
energy from the oscillations,but this absorptionis very small compared
to the energy that is delivered by the electrons in the spoke. The final
result is that the oscillations that are set, receive a substantial amount
of net energy from the electrons.
It may be interesting to note that, a cavity magnetron described
above is called the traveling wave magnetron precisely because of the
rotating RF fields.
Resonant Modes in Magnetrons : Resonant system of a mag-
netron oscillator consists of the anode cavities, together with the spaces
at the top and bottom of the anode block. The fields associated with the
cavities are such that the alternating magnetic flux lines pass through
the cavities parallel to the cathode axis and the alternating electrical
fields are basically confined to the slot and the region where the cavities
link the interaction space.
The number of resonant frequencies, or modes as they are called,
equals the number of cavities. This is because of the fact that resonant
system of the magnetron consists of a number of individual resonators,
one for each cavity, which are coupled together into one system. 'N'
number of cavity resonators give rise to an equal num5er of possible
modes.
The most commonly used mode in a magnetron is the a-mode. For
an eight cavity magnetron, a-mode corresponds to a total phase shift of
f 8n radians around the periphery. This n-mode corresponds to a single
phase system whereas other modes corresponds to polyphase systems.
The phase differences, that are characteristic of the various modes,
arise from the fact that each mode has a different frequency, and hence

Fig.5.21 Fig.5.22
is detuned differently from resonant fkequency of the cavities. Fig. 5.21
shows a relationship between the frequencies of different modes in a
magnetron. It is clear from Fig.5.21that in case of a normal magnetron
MICROWAVE TUBES 197

labelled as unstrapped, the wavelengths of different modes differ very

slightly from adjacent modes. This situation introduces practical dif-
ficulties which makes it important to separate the dominant a-mode
from other modes.
Strapping is a technique of separating K-modefrom other modes.
Strapping in its most elementary form consists of two rings of heavy
gauge wire connecting alternate anode poles. These poles should be in
phase with each other for the a-mode. Strapping achieves mode separa-
tion and prevents modes other than the a-mode from sustainingbecause
the phase difference between alternative anode poles will be different
from 25r: radians for all other modes.
Strapping may become unsatisfactory in high power magnetrons
because of losses in the straps or due to strapping difficulties at very
high frequencies. In case of high frequencies, the cavities are small and
large in number to ensure that a suitable RP field is maintained in the
interaction space.
Another method consists of an anode block with a pair of cavity
systems of very dissimilar shape and resonant frequency. One such
structure called a rising sun structure is shown in Fig. 5.22 and it has
the effect of separating the a-mode frequency from the others so that
the magnetron is not likely to oscillate at any of the other modes because
the DC field would not support them. Magnetrons with such a rising
sun cavity structure do not need any strapping.
Frequency Pushing and Pulling : The resonant frequency of
magnetron can be changed by changing the anode voltage. This frequen-
cy pushing, as it is called, is due to the fact that the change in anode
voltage results in a change in orbital velocity of electrons. This alters
the rate at which the energy is given up to anode resonators which then
changes the oscillating frequency.
Also, magnetron is susceptible to frequency variations due to
changes of load impedances. This takes place regardless of whether
these load variations are purely resistive or involve reactive load
variations. However, it is more severe in case of reactive variations.
This frequency variation is known as frequency pulling and is caused
by variations in load impedance reflected into cavity resonators.
Frequency pulling phenomenon is undesirable and should there-
fore be prevented.
5.6. Cross Field Amplifier
The Cross Field Amplifier (CFA)is basically the amplifier version
of magnetron oscillator and is therefore similar to it in construction and
operation. The important difference between the two is that the anode
slow wave structure of CFA does not close on itself. The two ends of this
slow wave structure are connected to separate transmission lines, one
for application of input and other for taking out the output as shown in
Fig. 5.23.As in case of a magnetron oscillator, there is a radial electric
field and an axial magnetic field. Current spokes are formed. These
198 MICROWAVES AND RADAR

Fig. 5.23
spokes rotate in a ciockwise direction around the cathode. Now, there
are two possible modes of interaction namely the forward wave inter-
action and backward wave interaction. With reference to Fig. 5.23, in
case of forward wave interaction, RF signal is applied at BB' terminals.
The signal flows around the circuit in clockwise direction and the
amplified output appears at A Y termnals. The current spokes travel
in synchronism with the circuit wave thus allowing it to grow.
The circuit shown in Fig. 5.23 is basically configured for backward
wave interaction. Here RF signal is applied atAA' terminals and taken
out at BB' terminals. The circuit comprises of a two wire line with
connections made to alternate vanes as shown. Though the circuit is
non re-enterent, the beam is re-entrant. As a result, the CFA could
oscillate if loop gain were high enough. This limits the gain to about 15
dB in amplifiers having a circuit completely around the circumference.
In case of some CFAs, the circuit encompasses only a fraction of
the total circumference as shown in Fig. 5.24 resulting in a significant
drifi space for the electron beam between the two ends of the circuit.
This type of CFA is capable of operating at wider bandwidth though at
a reduced electronic efficiency.
~ a c current
h spoke induces two circuit waves. One travels towards
the input and the other travels towards the output. The waves traveling
towards output add in phase while those traveling towards input cancel
each other. Since each spoke induces equal power into the circuit, the
wave grows with power increasing linearly with distance from input to
output. It is also important to not that with the wave growth being
MICROWAVE TUBES 199

linear, the circuit length would need to be increased by a factor of 10 if

Dielectric
filled wave

Fig. 5.24
i t was desired to increase the power gain by 10 dB. This is one of the
reasons why CFAs typically have low gain. Another point worth men-
tioning here is that the total power generated in a given CFA is
independent of RF input power as long as it exceeds the threshold value
required for spoke stability a t the input. The generated power can be
increased only by increasing the anode voltage and current. Fig. 5.25
shows output power versus input power characteristics with anode
voltage as a parameter. The input and generated powers marked in Fig.
5.25 are for an anode voltage of VAI.
The relevant expressions are :
Gain=(
Pin + P en
ping- ]=(I+%)
Overall efficiency,q = q e . q c

where qe = Electronic efficiency = --

[$:;A

and qc = Circuit efficiency =

200 MICROWAVES AND RADAR

n-
R F input Power

Fig. 5.25

Fig. 5.26
MICROWAVE TUBES 201

Cross field amplifiers offer a compact, high efficiency solution for

microwave amplification with relatively low voltage operation and
fractional bandwidth of the order of 10 percent. The efficiency figure is
significantly higher than traveling wave and klystron amplifiers. Cold
cathode operation is also possible eliminating the necessity of a high
voltage insulated heater power supply. Low phase pushing is another
attribute. The disadvantages include high noise level, typical of all cross
field tubes, low gain which necessitates a klystron or TWT amplifier
driver stage and its incapability to successfully handle amplitude
modulated signals due to it being a saturated amplifier. Fig. 5.26 shows
the photograph of a typical CFA capable of providing an average output
power of 15 kW, peak output power of 3 MW at an efficiency of about
80%and an operating frequency of 3GHz over a bandwidth of 200 MHz.
5.7. Travelling Wave Tube
The Travelling Wave Tube is an amplifier which makes use of a
distributed interaction between an electron beam and a traveling wave.
To prolong the interaction between an electron beam and an RF field,
it is necessary to ensure that they are both traveling in the same
direction with nearly the same velocity. It differs from the multicavity
klystron, in which the electron beam travels but the RF field remains
stationary. The electron beam travels with a velocity governed by the
anode voltage, which is typically 2 to 10%of the velocity of eledromag-
netic waves in free space. The interaction between RF field and moving
electrons will take place only when the velocity of RF field is retarded
by some means. This is achieved by wnat is known as a slow wave
structure. The slow-wave structures employed to retard R F field, either
use a helix or a waveguide arrangement.
The physical construction of a typical traveling wave tube using a
helix is shown in Fig. 5.27. It has an electron gun used to produce a
narrow electron beam, which in turn is passed through the center of a
long axial helix. A magnetic focusing field is provided to prevent the
beam from spreading and to guide it through the centre of the helix.
The signal to be amplified is applied to the end of the helix adjacent
to the electron gun. The amplified signal appears a t the output or other
end of the helix under appropriate operating conditions. Fig. 5.28 shows
the photograph of a typical TWT amplifier using a helix slow wave
structure.
Input Output
guide gu~de

Fig. 5.27
202 MICROWAVES AND RADAR

I
Fig. 5.28.
MICROWAVE TUBES 203

Operational Principle
The RF signal propagating around the turns of the helix produces
an electric field at the centre of the helix. Though RF field propagates
with velocity of light, the axial electric field due to RF signal advances
with velocity of light multiplied by the ratio of helix pitch to helix
circumference. When the velocity of electron beam travelling through
the helix approximates the rate of advance of the axial field, interaction
takes dace between the two.
The interaction is such that on a n average the electrons deliver
energy to the wave on the helix. As a result, the signal wave grows and
amplified output is obtained. The traveling wave tube may be thought
of as the limiting case of a multicavity klystron which has a very large
number of closely spaced gaps with a phase change that progresses from
left to right at approximately the same velocity as the beam.
The mechanism by which electroils transfer their energy to the RF
signal traveling through the slow wave structure is as follows. As
mentioned earlier, the operational principle of a traveling wave tube is
similar to a multi-cavity klystron with the difference that in case of a
TWT, the interaction between the electron beam and the RF signal is
continuous. The velocity nlodulation of electrons leads to the density
modulation due to non-zero trmsit time. Due to continuous interaction
between electrons and the RF wave moving through the slow wave
structure, the electrons get accelerated axialy. As these electrons
proceed along the tube, they bunch with those ahead of them. At the
same point along the circuit, after a half cycle of RF signal later in time,
the electrons are de-accelerated. These electrons tend to bunch with
those behind them. Now, the average velocity of electrons in the electron
beam is chosen to be slightly greater than the velocity of energy on the
slow wave circuit. Due to this, the electron bunches drift into a de-ac-
celerating RF field thereby delivering energy to the field. In this way,
the energy in the electron beam is converted into RF energy and
delivered to the slow wave circuit. Due to continuous distributed inter-
action, the RF wave grows'exponentially as it travels along the tube.
The purpose of using an attenuator is to prevent internal feedback
arising out of r~flectionsdue to impedance mismatch thereby leading
to generation of spurious signals.
The problem can be solved by placing an attenuator at some
convenient place in the tube. The glass wall of the tube coated with
aquadag acts as an attenuator. The attenuator absorbs the spurious
signals and to some extent absorbs growing signal in the helix. When
the attenuator is placed near the input end of the tube, the bunching of
the electrons remains unaffected. The reason for keeping attenuator at
the input end of the tube is that the attenuator attenuates both forward
and reverse waves, but the latter are more heavily attenuated than the
former.
Slow Wave Structure
The two commonly used slow wave structures include the Helix
and the Coupled Cavity circuit. The biggest advantage of a helix slow
 204 MICROWAVES AND RADAR
wave structure is the enormous bandwidth that it is capable of provid-
ing to the TWT. But it has its own problems when it comes to its use at
high frequencies and high power levels. The close proximity of helix
turns gives rise to spurious oscillations due to feedback at high frequen-
cies. Also, working at higher frequencies necessitates a smaller
diameter helix which is turn presents focusingdifficulties. Smallerhelix
diameter also prevents working at higher power levels as there is a
danger of this high power being intercepted by the smaller diameter
helix. Coupled cavity slow wave circuit (Fig. 5.29), on the other hand,
is well suited to high power, high frequency TWTs.

Fig.. 5.29

It comprises of a large number of overcoupled cavities with a

continuous phase shift progressing along the adjoining cavities. Such a
structure can be used to a maximum of 100 GHz. The bandwidth offered
by this type of slow wave structure is however no match to the excellent
bandwidth performance of a helix TWT.It is typically 10%of the center
frequency in case of a coupled cavity TWT as compared to as large as
100 % of center frequency in case of a helix TWT.
5.8. Backward Wave Oscillator
Backward Wave Oscillator (BWO)is the name usually given to the
oscillator version of a traveling wave tube amplifier though the cross
field amplifiers are also sometimes operated in this mode. The TWT
backward wave oscillator has the advantage of an enormous frequency
tuning range that can be conveniently achieved by varying a voltage.
Helix is the commonly used slow wave structure. The operational
principle of a BWO is similar to that of TWT amplifier with the
difference that the electron beam transfers its energy, present in the
form of electron bunches and acquired due to the velocity modulation
caused by RF field propagating in the slow wave structure, to the
backward moving RF field. The backward moving RF field grows in
strength and is taken out from the electron gun end of the tube. The
collector end is terminated in a passive termination. Oscillations can
always be assumed to start due to some transient and the presence of
an imperfect termination. The backward wave is further aided by
absence of attenuator which is present in case of a TWTA. Fig, 5.30(a)
shows the physical attributes of a BWO. Fig. 5.30(b)shows a photograph
of a typical BWO.
MICROWAVE TUBES 205

signal Termination
Electron
9'" , "Y"c

(b)
Fig. 5.30
To increase the interaction, BWO uses a ring cathode that emits
a hollow beam having maximumintensity near the helix. The oscillation
frequency primarily depends upon beam velocity which is turn depends
upon helix voltage and the associated cavity arrangement. Amplitude
of oscillations is governed by beam current which can be controlled by
varying the electron gun anode voltage.
Backward wave oscillators are widely used as voltage controlled
microwave oscillators because of the following features :
1.BWO can be tuned rapidly over a very wide range of frequencies
by changing voltage. The tuning range is as large as 100 percent (i.e.
one octave) upto 50 GHz which reduces to about half octive at higher
206 MICROWAVES AND RADAR
frequencies primarily due to use of multiple helixes or coupled cavity
slow wave structures.
2. BWO produces a very clean signal with the desired output signal
being ai least 60 dB larger than the total power of all spurious signals.
3. The frequency of oscillation is 2xLremely stable
4. BWO can be used at f.-equencies beyond 1000 GHz
5.9 Gyrotron
The size or dimensions of the interaction structure, where the
microwave field couples with the electron beam, in case of klystrons,
magnetrons, TWTs and &heirassociates namely Reflex klystron, CFA,
and Backward wave oscillatoi- becomes smaller as the frequency of
operation increases. Infhct, it is inversely proportional to the frequency.
These tubes become almost unusable a t 30 GHz and beyond. As an
example, a klystron tube that can handle about 1MW of average power
a t 3 GHz, can handle a mere 5 kW a t 30 GHz according to rough
estimates. It is because of the reason: that at frequencies of 30 GHz
and beyond, it is very difficult to fabricate the complicated interaction
structures that would allow the microwaves to travel a t electron beam
velocity. Even if they could be fabricated, these small interaction
structures would not be able to handle high average power.
The operation of a Gyrotron is based on the interaction of a static
magnetic field and the moving electr~nbeam. The moving electrons
traverse a spiralling path as they advance due to the effect of magnetic
field as shown in Fig. 5.31. The spiraling frequency called the cyclotron
frequency depends upon the strecgth of the magnetic field, mass of
electron and the velocity componen t perpendicular to the magnetic field.
Theory of relativity suggests that a fast moving object has higher energy
and therefore larger mass with the result that -cyclotron frequency
decreases. Also, energy coupling causes the electron to lose energy
which reduces its mzss with the result that cyclotron frequency in-
creases. All this leads to electron bunching similar to the one ex-
perienced in case of klystrons and TWTs.

Microwave I / P Microwave O / P

input cavity Center output cavity /;am

cuvlry collector
Magnet

Fig. 5.31
MICROWAVE TUBES 207
Fig. 5.31 depicts constructional features of a gyrotron tube. The
gun and the magnet are so designed as to have a gyrating path for the
moving electrons. The cavities are designed to have a beam opening ten
times larger than beam opening in a n equivalent klystron cavity in
order to accommodate the large diameter hollow beam.
When internal feedback is added to a gyrotron amplifier tube, it
becomes an oscillator. Gyrotrons when used as oscillators are the
highest power microwave oscillators a t frequencies of 100 GHz and
beyond. Gyrotron oscillators have produced 100 kW of CW power a t 100
GHz and more than 1 kW of CW power a t 300 GHz. These high
frequency oscillators used such a large strength magnetic fields that
the solenoids used for the purpose need to be cryogenically cooled.
5.10. Comparison of Microwave Tubes
I t is observed that some microwave tubes are better than the
others in one characteristic while others have a n edge in another
characteristic. A brief conlparison of different nlicrowave amplifier
tubes (Gridded tube, Klystron, TWT, CFA, Gyrotron) as regards
Bandwidth, Efficiency, Gain, Relative spurious signal level, Relative
operating voitage and Relative complexity of operation aye given in
Table 5.1. Average power and peak power handling capabilities of
different tubes as a function of frequency are shown in the graphs of
Figs. 5.32 ( a )and (b) respectively.
Table 5.1

Band- Effici- Gain Relative Relative Relative

width ency (dB) Spurious operating Comp-
(W (W Signal Voltage lexity of
Level * Opera
tion
X

Gridded
tube
1to 10 20 to 50 6 to 15 2
I Low / 1

Klystron 1to 5 30 to 70 40 to 60 ' 1 High 2

Helix 30 to 120 20 to 40 30 to 50 3 High 3
TWT
TWT 3 High 3
(coupled
cavity) I I I I I

Cross 5 to 40 1 40 to 80 ) 10 to 15 / 5 I Low 1 4
Field
Arnplifie~
(CFA)
Gyrotron 1 20 to 40 30 to 40 4 High 5
 MICROWAVES AND RADAR
Table 5.2

Type Mecha- Elect- Efi Relative Relative Relative

nical ronic cezncy Spurious Opera Comp
Tuning Tuning (%) Signal ttng lezityof
Range Range Level Voltage opera-
(9%) (%) * tion*

Gridded 10 10 to 40 2 Low 1
tube
Reflex 10
klystron
Back -
ward
Wave
Oscillat-
or (BWO)

Eklystron
Fixed fre-I -
quency
magnetr
on
Co-axial
magnetr
20 - 4Oto6Ol 5 1 Low / 3 I
on
Voltage -
tuned
magnetr
on
Gyrotron -
* 1-Best, 5 . 'oorest
Table 5.2 gives comparison of different microwave oscillator tubes.
Most microwave oscillator tubes with the exception of magnetron and
gyrotron operate in the CW mode. The average power capability of
microwave oscillators is considerably less than that of microwave
amplifiers. The peak power capability, however, compares with th& of
amplifier tubes.
Problem 5.1 : A reflex klystron is operating at 100 GHz. If the
mode operating in the tube corresponds to integer n = 3, determine the
transit time of the electron in the repeller space.
Solution : The transit time (T) in cycles is given by
MICROWAVE TUBES 209

3 15
for n = 3 , T = 3 + - = -cycles
4 4
for f = lOOGHz, one cycle corresponds to l/lol = 10 ps
15
Therefore, T = -x 10 = 37.5 ps
4

Fig. 5.32
Problem 5.2 :The operating frequency ofa reflex klystron is 2 GHz.
Calculate the change i n frequency for a 2% change i n the repeller voltage
given that :
Repeller voltage = 2000 V
Accelerating voltage = 500 V
Space between exit of the gap and the repeller electrode = 2 cm
(Assume that the operation is for n = I )
Solution : Change in frequency as a functional of repeller voltage
is given by :

-~ -V -R
df 2nn - n / 2
where e = charge of electron = 1.6 x lo-'' C
m = mass of electron = 9.1 x l o 3 ' kg
 210 MICROWAV3S AND RADAR

Problem 5.3 : I12 a certain reflex klystron, for given values of

gap-repeller electrode spacing and accelerating potential, 5% change i n
repeller voltage produces a 1% change in operating frequency when the
ratio of repeller voltage (VR)to operating fi-equency (fl is 1 VoltlMHz.If
this ratio is changed to 1.5 Volts1MHz, what would be percentage changc
i n (f) for 5%change in (VR)2

7
Solution :-- [2m2_"./2) m
*=

df;f
which gives [dvR,vR)= (
V R 2n n - n/2
2m ) 2
4
-

where K = constant = r2~)

6/-

whichgives ~='[f]='
5 VR 5
\ /

(7)'
New value of - = 1.5 Volts/MHz
'
percentage change in (n 1 3
Therefore, = 1.5 K = 1.5 x - = 0.3 = -
Percentage change in (VR) 5 10
3
or Percentage change in (f) = -x Percentage change in V R
10

Problem 5.4 :A certain Cross Field Amplifier has the following

parameters :
Anode voltage, V a = 40KV
Anode current, lo = 15A
Input power, Pin = 40KW
Gain = 1OdB
MICROWAVE TUBES 211

Determine the electronic efficiency ( q e ) Also determine the output

power i f the overall efficie~tcyis 40%.

\
[
Solution : Gain = 1+ Pb.e"
pin /
where Pgen= Generated power
Now, for a gain of 10 dB, gain ratio = 10

( %)
Therefore, 1+ - = 10

or Pgen = 360 kW

Also, Electronic efficiency, q e =

(*)
= 0.6
Overall efficiency, q = q e . q c

where qc = Circuit efficiency =

That is, (Pout -Pin.) = 0.66 x 360

= 237.6 KW
Or Pout = 237.6 + 40 = 277.6 KW

Problem 5.5 :Input cavity of a two cavity klystron is 0.1 cm wide.

I f the frequency o f the applied RF signal is 1 GHz and the beam
acceleratingpotential is 2500 Volts, determine the number of cycles that
would elapse during the transit of beam through the input gap. Assume
that RF signal amplitude is negligible as compared to accelerating
potential.
Solution :The velocity ( u )at which the electron beam enters the
gap is given by :

where e = Charge of electron -= 1.6~10-"C

m = Mass of electron = 9.1 x lo3' kg
V a = Accelerating potential = 2500 V
212 MICROWAVES AND RADAR

Gap spacing = 0.1 cm = 10" m

lo3
Therefore, time spent in the gap = = 0.34 x 10-1°s
2.93 x lo7

Now one cycle of RF signal equals ( l / f ) = (1/109)= 1 ns

Therefore, number of cycles = 0.034 cycle

Problem 5.6 :Determine the following i n case of an 8-resonator

traveling wave magnetron.
( a ) Phase difference between adjacent resonators i n general
( b ) Number ofpossible modes of resonance
( c ) Number of useful modes of resonance
( d ) Value of integer ( n )for the most dominant mnde

Solution :Let N = Number of resonators = 8

n = Any integer
+
= Phase difference between adjacent resonators
2nn 2nn nn
( a ) +=-=-=-
N 8 4
( b )Number of possible modes of resonance = N = 8
N
( c )Number of useful modes of resonance = - = 4
2
( d ) Most dominant mode is the one for which phase difference
between adjacent resonators is (n)radians.
nn
Therefore, -= n
4

Problem 5.7 :A two cavity klystron is operating at the electron

gun anode potential of 1200 volts. Determine the required peak
amplitude ofinput RF signal at 10 GHz frequency to generate maximum
output given that the spacing between the two cavities is 5 cm and the
gap spacing in either cavity is 1 rnrn.
Solution :The condition for maximum output is :

i 4
where 2nn - - = Transit angle between the two cavities
MICROWAVE TUBES 213

Vi = Peak amplitude of RF input

Vo = Accelerating Potential
If (2') is the transit time between the cavities,
5 x 10-~
then T = -
vo
where vo = velocity of electrons = dF

Therefore, T = - - 2.5 x 10's = 2.5 ns

2 x lo7
Transit angle = wT = 2nf T = 2 x 3.14 x 10 x 10' x 2.5 x lo-'
= 157 radians

V l = 3'68 1200 = 28.12 volts

157

Problem 5.8 : A helical TWThas a circumference (of helix) topitch

(of helix) ratio of 10. Determine the anode voltage for which the TWT can
be operated for any useful gain.

Solution :For TWT to operate for any useful gain, the axial phase
velocity of the RF signal should approximately equal the electron beam
velocity.

Electron beam velocity = d F

-

= 0.593 x lo6 (1)

Also, axial phasevelocity of RF signal propagatingin the slow wave
structure is given
- by :
Pitch
Free space velocity x
Circumference
214 MICROWAVES AND RADAR
From equations (1)and (21,
0.593 x lo6 dK=3 x lo7

In practke, the electron beam velocity is kept slightly greater than

the a a i phase velocity of RF signal.
MICROWAVE TUBES 215

Review Questions
1. What are the major limitations of conventional vacuum tubes
when they are used at microwave frequencies ? How do the
microwave tubes making use of transit time effects overcome
some of these disadvantages ?
2. Name three important microwave tubes that can be used as
microwave amplifiers. Briefly describe the principle of opera-
tion of each one of them.
3. What are the advantages achieved by adding more cavities in
between the input and the output cavities in a klystron tube ?
With the help of relevant schematic and Applegate diagram,
briefly describe the bunching phenomenon in a two-cavity
klystron.
4. Briefly describe the constructional features and principle of
operation of a Reflex 'klystron. Which tube parameters decide
the oscillation frequency and amplitude ?
5. Briefly describe, &th the help of a sketch, the principle of
operation of a cavity magnetron oscillator. What is the
dominant mode of operation in such a device ? How do we
ensure operatiqn at the dominant mode ?
6. Why do we call magnetron a cross field device ? How does a
cross field amplifier (CFA) differ from a magnetron oscil-
lator ? Briefly outline the principle of operation of a CFA.
7. Compare the undermentioned microwave amplifier tubes as
regards constructional and operational features.
(a) Multicavity klystron
(b) Traveling Wave Tube
(c) Cross Field Amplifier
8. Compare the salient features of the undermentioned
microwave oscillator tubes
(a) Reflex klystron
(b) Cavity magnetron
(c) Backward wave Oscillator (BWO)
9. Explain how a gyrotron tube has an edge over other microwave
tubes when i t comes to achieving high output powers a't high
operating frequencies.
10. What is the purpose of using a slow wave structure in a TWT
? Briefly describe the advantages and disadvantages of helix
type and coupled cavity type slow wave structures.
11. Explain the following phenomena :
(J) Velocity modulation
( b ) Bunching of electrons
Briefly describe the relevance of Applegate diagram to the
underst:ding of above processes.
12. What are the major characteristic features you would be
looking forward to have when you have chosen to use :
216 MICROWAVES AND RADAR

(a) a TWT amplifier with helix slow wave structure

( b ) a helical TWT based Backward Wave Oscillator
(c) a Gyrotron tube

Problems
The transit time of the electron in the reflector space of a reflex
klystron is observed to be 225 ps. Determine if this reflex klystron can
possibly oscillator at 10 GHz.
(No)
2. A reflex klystron oscillator operating at 10 GHz and a certain
repeller electrodevoltage shows a percentage change of 1%in frequency
for a 5%change in the repeller voltage. At another repeller voltage, the
same percentage change in repeller voltage causes a 1.5% change in
operating frequency. Determine the ratio of new to previous repeller
voltages.
(1.5)
3. A Cross field amplifier operates at an output circuit efficiency
of 60%. if the anode voltage and current are 50 KV and 20 A respectively
andif the power generated by the tube is 500 kW,determine the overall
efficiency.
(0.3)
4. Compute the electron beam velocity for a DC beam voltage of 10
KV.What will be the transit time (in cycles) for this beam to travel
through a 1mm cavity gap ?Assume RF signal frequency to be a 1GHz.
(0.593 x 10' m/s, 59.5 cycles)

5. The linear helix of a helical TWT has 20 turns per cm. Its
diameter is 5 rnm. Determine the axial phase velocity of the RF signal
of 10 GHz prapagating through the helix.
(9.554 x lo6 m/s)

SELF EVALUATION EXERCISE

I. Multichoice Questions
1. The most important factor that limits the frequency of opera-
tion of conventional triodes and tetrodes at microwave fre-
auencies is
(a) the electron transit time
( b ) the increased power losses
( c ) the limitation of the amount ofheat that can be dissipated
in the electrodes of conventional tubes
(d) the effect of interelectrode capacitances
MICROWAVE TUBES 217

( e ) the effect of interelectrode space charge

2. One of the following nlicrowave tubes is based on the principle
of velocity modulation
( a )Disk seal tube ( b )Acron tube
(c) Klystron (d)Magnetron
3. The microwave tube with the highest peak power and average
power handling capability a t frequencies of 30 GHz and
beyond is the
( a )Gyrotron ( b )TWT
( c )Magnetron ( d )CFA
4. One of the following is a microwave oscillator
( a )TWT ( b )CFA
( c ) Reflex klystron ( d )Klystron
5. Electron bunching in a 2-cavity klystron occurs in
( a ) buncher cavity
( b ) catcher cavity
(c) free drift space between the two cavities
( d ) none of these
6 . Energy transfer from the electron bunches to the output cavity
is the maximum when the transit of electron bunches through
the cavity is accompanied by
( a ) positive maximum of RF signal at the output cavity
( b ) negative maximum of RF signal a t the output cavity
( c ) an electric field just enough to bring electron bunches to
rest at the farther side of the gap
( d ) none of these
7. Repeller electrode is associated with one of the following
micro- ave tubes
( a )Multicavity klystron
( b )Gyrotron
( c )Crossed Field Amplifier
( d )Reflex klystron
8. One of the following is a crossed field device
(a)Magnetron (b TWT
(c) Two cavity klystron ( d )Reflex klystron
9. A travelling wave magnetron having 'N' resonators with
adjacent resonators tightly coupled has
( a ) N possible modes of resonance
( b ) (2N- 1) possible modes of resonance
 MICROWAVES AND RADAR
( c ) N/2 possible modes of resonance
( d ) none of these
Phase difference between adjacent resonators in a n N-
resonator travelling wave magnetron is given by

(".I
( a ) - radians

(%)
( b ) - radians where (12) i s an i n t q e r

( c )2 m radians where (12) is a n integer

( d )/$)radians where ( n )is a n integer

Pick the odd one out

( a ) Two cavity klystron amplifier
( b ) Multicavity klystron amplifier
( c ) Travelling wave tube amplifier
( d ) Backward wave travelling wave tube
A travelling wave tube having a helix length of 40 cm and
operated a t 30 GHz has helix characteristic impedance of 100
R and beam current to beam voltage ratio of 3.2 x lo-''. Gain
of the TWT in d B is approximately
( a ) 10 ( b )28 ( c )36 (dl 50
Of the following microwave amplifier tubes, the one charac-
teristised by largest bandwidth is
( a )Klystron ( 6 )Helix TWT
( c ) CFA ( d )Disk seal tube
Of the following microwave amplifier tubes, the one having
the lowest bandwidth is
( a )Gyrotron ( b )Helix TWT
( c ) Klystron ( d )Coupled cavity TWT
Of the following microwave amplifier tubes, one with lowest
spurious signal level is
( a )Cross-field amplifier (CFA) ( b )Klystron
(c) Helix TWT ( d )Coupled cavity TWT
One of the following microwave oscillators can be electronical-
ly tuned over very wide bandwidths.
( a )Reflex klystron ( b )Voltage tuned magnetron
(c) Backward wave oscillator ( d )Gyrotron
MICROWAVE TUBES 219

Cross field amplifier (CFA)is a very close associate of

( a )Mngnetron ( b )Helix TW-T
(c) Multicavity klystron ( d )Coupled cavity TWT
Depth of modulation of a velocity modulated beam is given by
Kvi
( b )-
va

where (Vi)is peak amplitude of applied RF signal and (V,) is

the accelerating potential prior to the gap
a-mode of resonance in a travelling wave magnetron is char-
acterised by
N
( a )integer n = N (b)n=-
4
N
( c ) integer n = - ( d )integer n = ( 2 N - 1)
2
where (N)
is number of resonators
20. One of the following microwave tubes does not use resonant
cavities.
(a)%-cavityklystron ( b )Reflex klystron
( c )TWT ( d )Magnetron
11. Match the microwave tube given on the left with the cor-
responding interaction process given on the right
Type of Microwave Tube Interaction Process
1. Disk seal tube ( a )Velocity modulation with
resonant cavities
2. Klystron ( b ) Grid control of beam
current
3. Backward wave osc;llator (c) Crossed field
4. Cross field amplifier (d) Velocity modulation with
travelling wave struc-
ture
5. Gyrotron ( e ) Spiraling beam
5. Magnetron
7. Helix TWT
 Semiconductor
Microwave Devices
and Integrated Circuits
ust like the traditional gridded tubes, the semiconductor devices

J such as bipolar transistors, FETs etc. too seemed to be troubled

by similar problems like interelectrode capacitances, lead induc-
tances and transit time effects which had bothered the tubes
earlier. So, we had the new range of semiconductor microwave devices
that exploited the transit time effects to their advantage. There have
been significant improvements in devices belonging to bipolar and FET
categories and these devices are still in use at microwave frequencies.
Today, we have semiconductordevices for all microwave functions such
as microwave amplification, oscillation, switching, limiting, frequency
multiplication and so on. The need for miniaturisation and advances in
integrated circuit technology led to the development of microwave
integrated circuits (MICs), both hybrid MICs as well as monolithic
microwave integrated circuits (MMICs).The chapter begins with a brief
introduction to the limitations of bipolar transistors when used at
microwave frequencies. This is followed by a detailed description of
other semiconductor microwave devices and Microwave integrated
circuits.
The topics covered in this chapter include :
High frequency 6.11. T r a n s i t Time
Limitations of Devices
Bipolar Transis- 6.12. IMPATT Diode
tors 6.13. TRAPATT Diode
Varactor 6.14. BARRITT Diode
Step Reco- 6.15. TUNNETT Diode
very Diode 6.16. MITATT Diode
Parametric 6.17. Microwave FETs
Amplifier 6.18. Maser
PIN Diode 6.19. Laser
Point Contact 6.20. MICs
Diode 6.21. M a t e r i a l s F o r
Schottky Barrier Making MICs
Diode 6.22. T h i c k Film
Tunnel Diode Hybrid MICs
Backward Diode 6.23. ~ h i Film
n Hybrid
Gunn Diode MICs
6.24. MMICs
SEMICONDUCTOR MICROWAVE DEVICES 22 1

6.1. High Frequency Limitations of Bipolar

Transistors
High frequency performance of microwave amplifiers using bipolar
transistors gets limited because of the two main reasons described
earlier in case of tubes. These are :
1. Effects of inter-electrode capacitances and lead inductances
2. Transit time effects
We are all familiar with inter-electrode capacitances as applied to
bipolar transistors and these can be prominently seen in their high
frequency equivalent circuit. In fact, inter-electrode capacitances are a
much bigger nuisance in bipolar transistors than they were in case of
tubes. The reason is the following. These capacitances in case ofbipolar
transistors also depend upon the width of the depletion region at the
junction which in turn depends upon the bias. The inter-electrode
capacitances in case of tubes are not so bias dependent. This makes it
difficult for the designer to predict the behaviour of transistor at
microwave frequencies. The other problem is that due to inter-electrode
capacitances, both common emitter current gain (P) and common base
current gain (a)become complex in nature having acquired reactive
components.
Lead inductance problem is less severe as the semiconductor
devices are smaller in size and consequently lead lengths are also
smaller. The effects of lead inductances are further minimised by using
low inductance packages and special geometries.
The transit time effects are similar to what is experienced in case
of tubes. Though the distances to be travelled by charge carriers are
much smaller in case of transistors and one may expect shorter transit
times, but i t is not so. The advantage of having shorter distances is
counterbalanced by slower speeds of movement of charge carriers.
Nevertheless, transistors have an upper frequency limit that is greater
than that for tubes.
The transit time comprises of the emitter delay time suffered by
majority carriers, base region transit time for the injected carried
depending upon base region thickness and impurity distribution, col-
lector depletion region transit time depending upon width of depletion
layer and drift velocity and collector region delay time.

Fig.6.1
 22 2 MICROWAVES AND RADAR
In view of the above, while fabricating microwave transistors, one
needs to have small electrode areas to reduce inter-electrode capacitan-
ces and narrow active regions to reduce the transit times. In case of
bipolar transistors, this really means that one should have a very small
emitter junction and a very thin base region. Various types of
geometries are used by various matufacturers. The one that is com-
monly used is shown in Fig. 6.1. These have an upper frequency limit
well in excess of 20 GHz.
6.2.Varactor
All semiconductor junctior. diodes exhibit a junction capacitance
when reverse b i a s e ~ .The charge free depletion region of any PN
junction in a semiconlluctorw~denswith the application of reverse bias.
Since the surfaces of charge free depletion region represent effective
plates of the capacitance of the junction, application of reverse bias
causes reduction in the value of the capacitance. A semiconductor diode
that is designed to maximise the capwitance variation with applied
reverse bias is the varactor. Fig. 6.2 E hows the circuit symbol and the
capacitance versus reverse bias characteristics of a typical varactor
diode.

cA

Reverse Voltage

(a) (b)
Fig. 6.2
In a semiconductorjunction in which the doping on the N-side and
the P-side is uniform right upto the plane of the junction, capacitance
C varies as (V- where V is the applied reverse bias voltage. For
junctions with suitable doping profiles, the power of the reverse bias
voltage dependence of capacitance can be designed in the range of
(- 1/3) to (- 1). The two sides of the junction in a varactor diode are
doped in such a way that the desired voltage variation of the depletion
width and hence capacitance is obtained. Refer to Fig. 6.2(b). The
capacitance is maximum for zero bias and varies inversely with applied
reverse bias voltage. The useful reverse voltage swing is between the
reverse saturation point (maximum capacitance) and the point just
above avlanche (minimum capacitance). Conduction and avlanche are
the two conditions which limit the voltage swing and hence the
capacitance variation.
Fig. 6.3 shows the equivalent circuit of a varactor. Co here is the
voltage variable junction capacitance. For a silicon varactor, typical
SEMICONDUCTOR MICROWAVE DEVICES 223

variation is from 5 pF to 25 pF. Rb is the series resistance. The load

inductance Ls and the stray capacitance (between anode and cathode)
become noticeable at higher microwave frequencies. Co and Rb together
decide the resistive cut-off frequency given by :
fo = 1/ [2r&bCo]
which is also regarded as the figure-of-merit of the device. This cut-off
frequency is certainly not the maximum frequency up to which the
device can be operated with any useful purpose. Varactors are typically
used up to one tenth of this cut-off frequency. At operating frequencies

Fig. 6.3
greater than (fo/lO), the series resistances increases , partly due to
skin effect, to a level where the diode Q drops to render it unusable
particularly as a frequency multiplier or in a parametric amplifier.
Increase in Rb increases noise which reduces signal to noise ratio. It
also increases losses which reduces efficiency.
Gallium arsenide is the most commonly used semiconductor
material for fabricating varactor diodes due to their higher mobility of
charge carriers. Gallium arsenide varactors have a resistive cut-off
frequency approaching 1000 GHz.
Varactor diodes find extensive use in microwave frequency multi-
pliers, parametric amplifiers and as a tuning element.
Frequency Multiplication Using Varactors
The mechanism responsible for frequency multiplication in a
varactor has its origin in the fact that when an AC voltage is applied to
a non-linear resistance, the output current versus input voltage
relationship is of the form :

Co-efficients c, d..... are the co-efficients of non-linearity and become

smaller and smaller as we include higher order terms. In fact, co-effi-
cient 'd' and beyond are so small that 'v' will have to be prohibitively
large to make any impact on the output. Now, if we consider only the
first three terms, then the output will contain second harmonic or in
other words, the input signal frequency gets multiplied by 2. But such
a multiplier is not efficient. If the same concept is applied to a non-linear
impedance or more particularly to a non-linear reactance, it has been
proved that frequency multiplication can be achieved with efficiencies
approaching 100%).This makes use of a varsctor diode, which provides
224 MICROWAVES AND RADAR
non-linear capacitive reactau~e(as capacitance varies as a f~~nction of
voltage), as a frequency multiplier very attractive. Fig. 6.4 shows a
typical circuit. It is a frequency doubler built using a varactor. The
circuit is self explanatory.
Varactor diodes are particularly suited when broadband frequency
multiplication is desired. They are not suitable for larger n~ultiplication
factors where the preferred choice is the step recovery diodes, to be
discussed next. When compared to step recovery diodes, varactors can

Input
Matcher
Zf! varacier f 0~1-t
Matcher

Fig. 6.4
be used to much higher frequencies exceediug 100 GHz and have a
higher power handling capability. Step recovery diodes, on the other
hand, are used up to about 20 GHz. In a practical frequency multiplier
circuit, where a large multiplication factor as well as a higher power
handling capability is desired, one tends to use a chain of step recovery
diode frequency multipliers as well as varactor diode multipliers with
the former providing most of multiplication and the latter providing
desired output power level at the required frequency.

6.3. Step Recovery Diode

Step recovery diode, similar in construction to a varactor diode and
fabricated from Silicon or Gallium Arsenide is mainly used as a
microwave frequency multiplier. The principle of operation of a step
recovery diode is however different from that of a varactor diode. A step
recovery diode stores energy in the junction capacitance when it is
forward biased and conducting and then releases it in the form of a
sharp current pulse when reverse biased. The diode is so biased that it
conducts during a part of the input cycle and it is during this period
that its depletion layer is charged. When the signal reverses polarity,
the diode is reverse biased and the stored energy is released in the form
of a sharp current pulse that is very rich in harmonics. A tuned circuit
placed in the output and tuned to the desired harmonic frequency is all
that is required to achieve frequency multiplication. In a step recovery
diode based frequency multiplier, i t is important that the discharge
current pulse width is less than the reciprocal of the desired output.
frequency. Conversion efficiency well in excess of reciprocal of multi-
plication factor is easily achievable with step recovery diodes. Another
common application of step recovery diodes is in comb generators to
produce multiple harmonics.
SEMICONDUCTOR MICROWAVE DEVICES 225

As mentioned earlier, step recovery diodes are the preferred lot

when it comes to achieving higher multiplication factors a t relatively
higher efficiency values than varactor diodes. Multiplication factors as
high as 35 have been reported in the literature. Efficiency can be as
high as 80 percent at lower end of microwave frequency band.

6.4. Parametric Amplifiers

A Parametric Amplifier is not a device. Instead, it is a circuit used
to provide low noise amplification a t microwave frequencies. It has been
chosen to be discussed here as it invariably uses a varactor diode to
provide low noise amplification. It is based on the principle of electroni-
cally varying the capacitive reactance (which is a parameter of the
tuned circuit and hence the name parametric amplifier) of a device,
usually a varactor, at a frequency higher than the frequency of the
microwave signal to be amplified. The signal used to vary the capacitive
reactance of the signal electronically is called the pump signal and the
amplification mechanism is provided by a varying reactance. This is in
constrast to the use of a DC power supply and a variable resistance in
case of conventional amplifiers. It is this novel amplifying mechanism
using a variable reactance rather than a variable resistance that is
responsible for making this amplification process a low noise one. In a
parametric amplifier, usually three frequencies are present namely the
Signal Frequency 'fs', the Pump Frequency 'fp'and the difference of the
two frequencies called the Idler Frequency 'fi'. The amplified output is
taken either a t the signal frequency or at the idler frequency which
could be lower or higher than the signal frequency depending upon the
frequency of the pump oscillator. Irrespective of the mode of operation,
the output gets amplified a t the expense of the energy drawn from the
pump source.
Parametric amplifiers found widespread early acceptance as a
highly sensitive, low noise microwave amplifier to be used as a front
end in a microwave receiver as an alternative to the arrangement of a
silicon point contact diode mixer followed by an IF amplifier. Radio
astronomy that required extremely high sensitivity microwave recep-
tion provided the first major application for parametric amplifiers.
Though the parametric amplifiers (called paramps in short) had an
added complexity, yet they found wide acceptance because of the sig-
nificant improvement in sensitivity they offered. Also, they had a very
good noise performance which could be further improved by operating
the paramp at low temperature. These have been operated a t cryogenic
temperatures such as liquid Nitrogen and liquid Helium temperatures
in some special applications where extremely low noise performance is
desired. Recently, however, these have been replaced by Gallium Ar-
senide MESFET (discussed in the latter part of the chapter) amplifiers
as they offer comparable noise performance with reduced complexity.
Degenerate Parametric Amplifiers
Having introduced the parametric amplifier, we shall now discuss
different types of paramps, their operational mods and characteristic
226 MKROWAVES AND RADAR
features. There are two broad categories of paramps name:y Degenerate
Paramps and Non-degenerate Paramps. In degenerate paramps, the
pump frequency is approximately twice the signal frequency with the
result that the idler frequency (fi = fp - fs) approximately equals the
signal frequency. When the pump frequency exactly equals twice the
signal frequency, the signal frequency and idler frequency bands com-
pletely overlap each other. Such an amplifier is characterised by a
negative resistance i.e. it presents a negative resistance to an external
circuit at the signal frequency. The output is taken a t the signal
frequency. Fig. 6.5 shows the basic amplifier circuit of the degenerate
type where the pump frequency is exactly twice the signal frequency.
Such an amplifier has the following drawbacksflimitations :

Fia.
- 6.5
1. Its output contains noise contributions from both the signal as
well as idler bands as the two bands would overlap.
The configuration shown in Fig. 6.5 being a 1-port one neces-
sitates that the input and output circuits are isolated. The
isolation can be achieved by using some isolation component
such as a ferrite circulator. In the absence o f any isolation, if
such a paramp was followed by a relatively noisy stage such
as a mixer, the mixer noise will find its way to paramp input,
get amplified with the result that noise performance would
suffer. Fig. 6.6 shows the use of a ferrite circulator in isolating
the input of an paramp, usually fed from a microwave antenna,
from the output feeding the mixer input. ks is clear from the
figure, mixer noise can never find its way either to the paramp
or the antenna. It gets routed to the matched termination.
Such an arrangement provides adequate isolation. Some of the
recent designs employ five port circulators to provide in-
creased immunity from the effects of source impedance varia-
tions and to provide greater isolation.

The one port paramp ofthe type shownin Fig. 6.5 where the output
is extracted at the signal frequency could be a nondegenerate one too
if the pump frequency is not equal to twice the signal frequency.
Non-degenerate paramps are described in the following paragraphs.
SEMICONDUCTOR MICROWAVE DEVICES 227

henna7

Signal -
IN

--
+

Fig. 6.6
Non-DegenerateParametric Amplifiers
In a nondegenerate type of parametric amplifier, the signal and
idler bands are clearly separated i.e.the pump frequency 'fi' is not equal
to twice the signal frequency 7s'. One of the varieties of a non-
degenerate paramp is a one port configuration shown in Fig. 6.5 where
the output is taken at the signal frequency. But in most of the non-
degenerate designs, the output is taken a t the idler frequency
(fi = fi - fs). Fig. 6.7 shows the basic configuration of this type.

Fig. 6.7
As is evident from the figure, this amplifier codguration is a two
port one and thus does not require an isolator for isolating input and
output. There are two commonly used non-degenerate amplifier con-
figurations. The first type is the Up-Converter where pump frequency
7P7is much higher than the signal frequency 'fs', usually 5 to 10 times,
so that idler output, which is also the amplified output, is larger in
frequency than the signal frequency. The other type is the Down-Con-
verter where the pump frequency 'fi>' is only slightly larger than the
signal frequency 'fs' so that the difference of the two is lower than the
signal frequency.
228 MICROWAVES AND RADAR
The non-degenerate amplifier does not have any compulsion of
having a pump frequency that is a multiple of the signal frequency. It
is a more versatile configuration than its degenerate counterpart as it
can be used to extract amplified output at either the signal frequency
or a frequency that is much higher than the signal frequency or even
a t a frequency lower than the signal frequency. Like a degenerate
amplifier configuration, this too offers a negative resistance.
Yet another paramp configuration that offers a positive resistance
is the one where the output is taken a t a frequency equal to ( f p + fi). It
has low gain but has the advantage that it requires a lower 'fp' and is
attractive while working at higher microwave frequencies.
Broadband Parametric Amplifiers
The tuned circuits used in the paramp configurations described
above are either of cavity or co-&a1 resonator type with the result that
these amplifiers offer a very narrow bandwidth, ra-sly exceeding 10
percent of the center frequency due to a very high Q-factor associated
with these tuned circuits. One of the useful methods to enhance the
bandwidth is to use a multiple stage low pass filter with multiple
inductors either being lumped components or being provided by a
transmission line structure and capacitances provided by varactor
diodes. Fig. 6.8 shows the basic arrangement of a paramp using such a
traveling wave structure. The bandwidth of the low pass filter should
be such that the signal frequency 'fs', the pump frequency '/j~' and the
idler frequency 'fl lie within the pass band and the sum component
'& + fs' lies outside the passband. Meeting both the conditions puts an

Fig.6.8
upper limit on the maximum usable pump frequency. If the pump
frequency is much higher than the signal frequency, then the sum
component is not much different from the difference component and the
desired filtering becomes difficult to achieve. The signal and pump
inputs are applied a t the input of the multiple low pass arrangement
and the signal output and idler components are taken from the output
of this multiple LPF. The output is suitably terminated so that there
are no reflections at the desired output frequency.
6.5. PIN DIODE
A PIN diode consists of a heavily doped P-type semiconductor
material (P+)and a heavily doped N-type semiconductor material (N+)
separated by a layer of extremely high resistivity intrinsic semiconduc-
SEMICONDUCTOR MICROWAVE DEVICES 229

tor material. Fig. 6.9 shows the cross-sectional view of a typical PIN
diode. In practice, no semiconductor material is absolutely free from
impurities. A more practical definition of a PIN diode calls it a semicon-
ductor diode which consists of two heavily doped P and N regions
separated by substantially higher resistivity P or N region. This leads
to two types of PIN diode structures. The first is the one of heavily doped
P and N regions separated by an unusually lightly doped P-type
intrinsic layer (alsoreferred to as n-type) and the other is heavily doped
P and N regions and a layer of lightly doped N-type intrinsic material
(also referred t o as v-type) sandwiched between the two. Except that in

I Layer

Fig. 6.9
case of the former the semiconductorjunction occurs at the N+interface
and in case of the latter, it occurs at P+interface, there are no other
major performance differences.
Operational Basics
When a practical PIN diode is unbiased (zero bias), a small region
around the junction will be depleted of charge carriers and since I-layer
is verylightly doped, a major portion of I-layer will be depleted and the
depletion zone extends very little into the heavily doped semiconductor
region. Fig. 6.10 shows the phenomenon in case of a PIN diode using a
v-type I-layer. Fig. 6.10(b)shows the ionised impurity profile for a zero
bias. As the reverse bias voltage is increased, the depletion zone extends
rapidly into the I-layer due to its light doing level and very slowly into
the P+region. At a sufficient reverse bias level equal to Punch Through
Voltage, I-layer is completely depleted of charge carriers as shown in
Fig. 6.10(c)producing a nearly open circuit across the device terminals.
The other relevant parameter with the PIN diode under reverse bias is
the junction capacitance. The junction capacitance decreases as the
reverse bias increases due to widening of the depletion zone. It reaches
its minimum value at the punch through voltage beyond which it
remains constant. Fig. 6.11 shows the typical Junction capacitance
versus Reverse bias voltage characteristics of the PIN diode. As is clear
from the curve, the characteristics have a soft knee. The punch through
voltage, by convention, is the voltage correspondingto two straight line
projected slopes as shown. It may be mentioned here that such a
Capacitance versus Voltage characteristics would be obtained if the
measurements are done at a relatively low frequency, typically 1MHz.
At microwave frequencies, the junction capacitance would be constant
and equal t o Cmin irrespective of the applied reverse bias voltage.
230 MICROWAVES AND RADAR

Reverse bias
f + + + +

Fig.6.10

Fig.6.11
This is due to the reason that silicon has a high dielectric constant
andits dielectric susceptabilityat microwavefrequenciesis much larger
than the conductivity of either v-type or x;-type intrinsic material. The
frequency at which the two equal is called dielectric relaxation frequen-
cy of the material. At an operating frequency equal to or more than
three times this frequency, the junctibn capacitance is constant at
C ~ ,irrespective of reverse bias voltage. The junction capacitance
offered by a PIN diode at microwave frequencies is particularly impor-
tant when it is to be used for microwave switching applications. The
PIN diode capacitance measured at a low frequency such as 1MHz to
a good approximationrepresents the effectivecapacitanceat microwave
frequencies provided that while making low frequency test, it should be
ensured that the supplied reverse bias fully depletes the I-layer.
SEMICONDUCTOR MICROWAVE DEVICES 23 1

Under forward bias, the PIN diode principally behaves as an

electrically variable resistor. When increasing forward bias is applied,
more and more charge carriers (Holes from P+ and Electrons from N+
regions) are injected into the I-layer thus decreasing its resistivity. This
conductivity modulation by a varying forward bias is made use of in its
application as a variable attenuator.
Equivalent Circuit
Fig. 6.12 shows the reverse biased and forward biased equivalent
circuits of a PIN diode. Referring to the reverse biased equivalent circuit
shown in Fig. 6.12(a),Rs represents the equivalent series resistance of
both ohmic contacts and undepleted silicon regions. Cp and Ls are
contributed by parasitics associated with device packaging. The value
of Rs ranges from fraction of an ohm to about 3 ohms. Since the
reactance associated with parasitic Lsis very small (Lsbeing a fraction
of a nano henry typically), the diode can be considered as a low loss
capacitor in the reverse biased mode particularly at lower microwave
frequencies. The junction capacitance, to a good approximation, can be
computed from :
Ci = [EOG A/Wl
where A = Area of I-layer
W = Thickness of I-layer
If D is the cross-sectional diameter of the I-layer,
CL= [&o&rdI2/4 UI]

6
(a) (b)
Fig. 6.12
In the forward biased mode, Rfrepresents overall device resistance
which has two components. One is a constant resistance (similar to
Rs in Fig. 6 . 1 2 ~accounting
) for mainly the ohmic contacts and the other
is a variable component due to the intrinsic layer. This variable com-
ponent Ri is a function of forward biased current and is given by :
Ri = [ w 2 / 2 . 1 ~ ~ 1 . ~ 1
232 MICROWAVES AND RADAR
where If=Forward biased current
p = Average carrier mobility
z = Carrier lifetime in I-region
It is in this operational mode that a PIN diode can be used as a
variable attenuator. If we ignore the parasitic components resulting
from device package and look at the equivalent circuit of the PIN diode
chip only, then we get the equivalent circuit of Fig. 6.13. As is clear from
the figure, the diode is simply a forward biased resistance Rf in the
forward biased mode and a series combination of Rr and Ci in the
reverse biased mode. The two equivalent circuits help us a great deal
in comparing the performance of a bare diode and a packaged diode and
the degradation in the performance specifications that occurs due to
device packaging. It may also be mentioned here that PIN diodes in chip

Fig. 6.13
form are also used in specific applications and also when they form a
part of an MMIC (Microwave Monolithic Integrated Circuit).
Performance Parameters
1. Dielectric Relaxation Frequency : Dielectric relaxation
frequency fR is the frequency at which the dielectric susceptability
equals the conductivity of the I-layer (rc-typeor v-type). It is an indica-
tive of the frequency at which the junction capacitance Cj becomes
constant equal Cmin .
2. Cut-Off Frequency : Cut-Off frequency fe is the upper fre-
quency limit of the diode. It is given by :
fc = [1/2rc. 'Cj.Rr]
3. Switching Cut-Off Frequency : It is slightly lower than the
cut-off frequency and is the effective cut-off frequency for a PIN diode
switching application.
fc (switching) = [1/2rc. Cj m fl
4. Transition Frequency :It indicates the lower limit. It is the
frequency at which the transit time of holes (slower of the two types of
carriers) through I-region equals the operating frequency time period.
SEMICONDUCTOR MICROWAVE DEVICES 233

f~= l / T p where Tp = Transit time for holes

Transit time further depends upon carrier mobility and is given
by :

where W = I-layer thickness

Dp = Diffusion constant
Also, Dp = [K.T/el.p
where K = Boltzmann's constant
T = Absolute temperature
e = Electronic charge
p = Carrier mobility (Drift velocity per unit electric field)

This gives f i = [ D ~ / w ~ ]
Other parameters of interest are the junction capacitance Cj, Forward
biased resistance Rf, Reverse biased resistance Rr and parasitic com-
ponents.
PIN Diode Applications
PIN diode as an Attenuator : PIN diodes can be used as a
variable attenuator when operated in the forward biased mode. Typi-
cally, the forward biased resistance can be varied from a few ohms to
several kilo-ohms for a variation in the forward biased current from a
few microamperes to tens of milliamperes.
PIN Diode Switch :PIN diode due to its behaviour almost as a
short circuit when sufficiently forward biased and a n open circuit when
zero or reverse biased and also due to its capability to rapidly swltch
between the two states makes it a very good microwave switch. It has
reasonably good power handling capability and several devices can be
connected in parallel to enhance i t further. Different configurations can
be used to simulate SPST (Single Pole Single Throw), SPDT (Single Pole
Double Throw) and SPMT (Single Pole Multiple Throw) switches. Fig.
6.14 shows some ofthe possible SPST configurations. Fig. 6.14(a)shows
a series switch configuration where the microwave power flows from
input to output when the PIN diode is forward biased. In case of the
configuration shown in fig. 6.14 ( b ) ,i t flows when the diode is open or
reverse biased. Figs 6.14 (c) and ( d )respectively show series-shunt and
T-connected configurations. In these two types, power is allowed or
disallowed to flow depending upon which diode is open or closed.
Fig. 6.15 shows one possible SPDT configuration. The microwave
power flows to the output port-1 when the driver produces a LOW a t
junction X and a HIGH a t the junction Y. The power flows from input
to the output port-2 when X and Y are respectively HIGH and LOW.
The reasons are obvious.
234 MICROWAVES AND RADAR

Transmission Line Transmission Llne

(Out)

Driver

(a)

Transmission Line
A 0B
(Out)

Driver

(out)
PIN-2

Fig. 6.14.
An SP4T switch simulation using PINdiodes is shown in Fig. 6.16.
The microwave power flows from input to an output port whose relevant
driver output is LOW. For instance, if the driver output marked A is
LOW and all other outputs are HIGH, PIN diodes Dl, D6,D7 and 0 8
are forward biased while diodes D2,D3,D4 and 0 5 are reverse biased.
Reverse biased diodes D2,D3 and 0 4 disallow the power to go to output
ports 2, 3, or 4.
SEMICONDUCTOR MICROWAVE DEVICES 235

Fig. 6.15

Driver
I

Fig. 6.16
PIN Diode Phase Shifter : Microwave phase shiftingis another
common application of PIN diodes. Infact, in the phase shifting applica-
tions too, PIN diodes are used as switches where its low resistance in
the forward biased mode and a low loss capacitancein the reverse biased
236 MICROWAVES AND RADAR
mode is made use of. There are four commonly used phase shifting
circuit topologies namely Loaded Line Phase Shifter, Reflection Type
Phase Shifter, High P a s s h w Pass filter Phase Shifter and Switched
Line Phase shifter.
In the Loaded Line Phase Shifter as shown in Fig. 6.17, a trans-
mission line with a characteristic impedance Z and electrical length
'8' is terminated in two identical admittances which can be switched
using PIN diodes. Advantages of loaded line phase shifters are low loss
and simplicity. Bandwidth constraints restrict their use upto a phase
shift of less than equal to 45".
The basic Reflection Type Phase Shifter, as shown in Fig. 6.18,
consists of a variable or switched impedance elements terminating the
coupled ports of a 3 dB hybrid coupler. The output is from the isolated
port.
out
Z, 8
T "

Fig. 6.17

Fig. 6.18
A Switched Line Phase Shifier as shown in Fig. 6.19 uses two
SPDT microwave switches to select alternate transmission line of
lengths 'l' and '1 + A I' to achieve phase shifting action. The phase shift
'A+' is given by :
A+ = p.Al
where p = Propagation constant
In the High Pass / Low Pass Filter Phase Shifter, phase shift is
accomplished by alternately switching in a high pass or low pass filter
section into the transmission path. At lower microwave frequencies,
lumped components can be used for filters. At millimetric wave frequen-
cies, diodes can be used. Different types are shown in Fig. 6.20.
SEMICONDUCTOR MICROWAVE DEVICES 237

In o o out
> P 7
I+AQ

Fig. 6.19

Fig. 6.20.

6.6. Point Contact Diode

A Point Contact Diode (Fig. 6.21) is one of the earliest diodes used
for microwave mixing and detection applications. I t has a metal-semi-
conductor junction unlike conventional PN junction diodes. Such a
device having metal semiconductorjunction is a majority carrier device
and is therefore free of all the ill effects of the minority carrier storage
associated with the devices using both majority and minority carriers.
Absence of minority carrier charge storage capacitance makes point
contact diode and Schottky barrier diodes (to be discussed in the next
section) particularly attractive, for mixer (Down Converter) applica-
tions, as varietions in capacitance are usually undesirable in a mixer
application. For microwave detection and mixing applications, it has
invariably been replaced by Schottky barrier diodes due to latter's lower
forward resistance, wide dynamic range and better noise performance.
238 MICROWAVES AND RADAR

Metal (Cathode)

Semiconductor

Fig.6.21
6.7. Schottky Barrier Diode
Schottky Barrier Diode, also called the Hot Carrier Diode, is by far
the most commonly used mixer and detector diode a t microwave and
millimeter wave frequencies. It is an extension of the concept of the
metal-semiconductor junction used to advantage in the point contact
diode discussed in the preceding paragraphs with the difference that
the metal-semiconductor interface in case of a schottky barrier diode is
a surface rather than a point as was the case in a point contact diode.
Like the point contact diode, it has tIte advantage that it is a majority
carrier device and there is no storage of minority carriers when the
device is reverse biased with the result that the delay associated with
removal of stored charge is absent. This leads to increased speed in case
of these devices.

,5102
Layer
N- type GaAs Active Layer
Buffer Laver

Ohm~cContact

Fig.6.22 Fig.6.23
Commonly used semiconductor materials for fabrication of schot-
tky barrier diodes are silicon ( S i ) and gallium arsenide (GaAs). While
gallium arsenide has higher cut-off frequency and lower noise, silicon
schottky barrier diodes are easier to fabricate. Practical devices are
available for use at frequencies exceeding 100 GHz. Fig. 6.22 shows the
cross-sectional view of a typical schottky barrier diode.
SEMICONDUCTOR MICROWAVE DEVICES 239

The equivalent circuit of a typical schottky diode is shown in Fig.

6.23. Rs represents losses in ohmic contacts, substrate and undepleted
epitaxial layer. Cj is the junction voltage dependent junction
capacitance and Rj is the current dependent junction resistance. Lp and
Cp are the parasitic components due to device packaging. The cut-off
frequency of the diode is given by :
fc = 1/2n Rs .Cj
6.8.Tunnel Diode
A Tunnel Diode, also known as Esaki diode after the name of its
inventor, is a two terminal device that offers high conductivity for both
forward as well as reverse bias and also exhibits negative resistance for
a very small applied forward bias. It lends itself very well to be dsed as
an amplifier or as an oscillator at microwave frequencies, though its use
as an amplifier is more common.
A tunnel diode is also a PN junction diode but with a significant
difference in the impurity concentration of the dopant. In the conven-
tional semiconductor diode, the impurity concentration is about 1part
in lo6 parts. The depletion width, which is inversely proportional to the
square root of impurity concentration, is about 5 microns. In case of a
tunnel diode, the impurity concentration is about 1part in lo3 parts
and the depletion width is approximately 0.01 micron. Thus the deple-
tion width is very narrow in case of a tunnel diode. It has been observed
that if the depletion region is too narrow as is the case in a tunnel diode,
the charge carriers instead of climbing up the potential barrier like they
do in case of a conventional PN junction diode, may pierce through the
junction with little or no bias applied. This phenomenon, called tunnel-
ing, is responsible for its behaviour as a very good conductor both in
forward as well as reverse direction for a very small applied bias of the
order of 0.1 volt. Fig. 6.24 shows the I-V characteristics of a tunnel
diode. From its characteristics, it is clear that both in the forward as
well as reverse directions, the diode responds with a huge current for a
very small applied voltage.

Fig.6.24
240 MICROWAVES AND RADAR
In the forward direction, the current reaches a maximum value
equal to Ip (called peak current) at a voltage V p (called peak voltage).
At this point di/dv is zero. Beyond this point, the current starts
decreasing with increase in applied voltage till another point (called
valley) is reached where again d i / d v is zero. After this, the current
starts increasing again with increase in voltage and a current Ip may
be encountered again at a voltage equal to Vfi At the peak point, the
slope of the characteristics changes from positive to negative while at
the valley point, i t changes from negative to positive again. Between
the points, the device exhibits negative resistance.
These characteristics may be considered to be composed of two
characteristics, one due to the tunneling action and the other due to
normal PN junction diode action as depicted in Fig. 6.25. Due to
tunneling phenomenon, the decrease in current beyond the peak point
may be justified on the basis that there are no more carriers available
for tunneling so that a reduction in current results with increase in
voltage. The current approaches zero for a forward voltage of about 0.4
to 0.5 volt. But then the normal PN junction effect starts. The current

I \ I dibde /

Fig.6.25
due to normal PN junction diode effect adds to the tunneling current to
give rise to a valley point as shown. The characteristics thus exhibit a
negative resistance from peak to valley points.
Tunnel diodes are constructed either from Germanium or Gallium
Arsenide. Silicon is not used. This is due to the reason that the ratio
( I p / I v )is maximum in Gallium Arsenide (= 15)and relatively smaller
for Germanium (= 8) but very small for Silicon (= 3). In addition to this,
Gallium arsenide offers maximum voltage swing (= Vf - V p ) of 1 volt. A
voltage swing of 0.45 volt is common with Germanium tunnel diodes.
Considering the above stated reasons, Gallium arsenide is the most
commonly used semiconductor material for making tunnel diodes.
Fig. 6.26 shows the small signal equivalent circuit of a tunnel
diode. R s represents the series resistance of the ohmic contacts and the
bulk semiconductor. Ls is the lead inductance. R and Cj are respective-
ly the negative differential resistance computable from the slope of the
negative resistance region o f its I-V characteristics a n d junction
SEMICONDUCTOR MICROWAVE DEVICES 24 1

capacitance. The tunnel diode exhibits negative resistance up to an upper

frequency limit given by :
fc = 1/2 IT.R.Cj
This is also known as the resistive cut-off frequency ofthe tunnel
diode or its figure-of-merit. It is the frequency at which the magnitude
of negative resistance equals the magnitude of the reactance offered by
the junction capacitance. The above expression ignores the effect of
parasitic components. Considering the effect ofRs, the same expression
can be written as :
where 'Rmin' is the minimum value of the negative differential resis-
tance.

Fig. 6.26
Tunnel diodes find application as microwave mixers, detectors, low
noise amplifiers and oscillators. It is a low noise device and its low noise
performance primarily arises from two reasons. These are its extremely
small series resistance Rs leading to a low thermal noise contribution
and its low current operation keeping the shot noise to a bare minimum.
However, the power output of tunnel diode amplifiers and oscillators is
limited to a few milliwatts and due to this reason, these are being fast
replaced by some of the newer microwave solidstate devices particularly
the Gallium arsenide FETs.
6.9. Backward Diode
Backward Diode is a special type of a tunnel diode. By varying the
doping concentration, the I-V characteristics of the device are so
modified to suppress its peak point with the resultant characteristics
as shown in Fig. 6.26. The doping concentration in this case is barely
large enough to cause tunneling. The device is operatedin the backward
bias region of its characteristics. For a small applied reverse bias, it
responds with a tunnelingcurrent that is much larger than the current
for a corresponding bias in the forward direction. The backward diode
has a very low (l/f)noise due to which i t makes a good choice as a
microwave mixer and detector diode. Due to absence of minority carrier
storage effect, the device has an excellent frequency response. As the
tunneling effect is predominant, the I-V characteristics are insensitive
to temperature changes unlike conventionaljunction diodes whose I-V
characteristics vary strongly with changes in operating temperature.
242 MICROWAVES AND RADAR

Fig. 6.27

6.10.GUNN Diode
When it comes to using semiconductor devices at high frequencies
approaching microwave frequencies and beyond, they encounter
problems whose cause is very different from that responsible for making
triodes, tetrodes and pentodes unusable at microwa~2 frequencies. And
the cause, discussed earlier with reference to bipolar transistors at
microwave frequencies is the transit time effects. In case of tubes, it
was overcome by developing a new class of tubes that depended for
their operation on the transit time rather than getting bogged down by
it. We all know microwave tubes namely the Klystron, the Magnetron
and so on. In case of semiconductor devices too, for successful operation
a t microwave frequencies, we need to either drastically reduce the
transit time for which device's junction needs to be made extremely thin
or make use of transit time phenomenon. Since the device junction can
not be made thin beyond a certain limit if it is to handle a certain
microwave power level, the other option is to use transit time to
advantage. Yet another option is to use the properties of the bulk
semiconductor material rather than its junction having its associated
transit time effects. A large number of microwave semiconductor
devices have been developed making use of the two above mentioned
effects. One class of devices which takes into account transit time in its
operation shall be discussed in the next section. The other class of
devices is the one which does not use the junction properties and
depends for its operation on the properties of the bulk semiconductor
material. The Transferred Electron Effect, also known as Gunn Effect
after the name of the scientist who initially discovered this effect is
responsible for producing microwave oscillations in the semiconductor
devices. This effect depends upon the properties of the bulk semicon-
ductor material and therefore the devices based on this effect do not
have a junction. Gunn diode (Transferred Electron Device) is the most
prominent member of this very small group of microwave devices and
it will be discussed in detail in the following paragraphs.
Transferred Electron Effect
According to transferred electron effect (or Gunn effect), the
electrons in a high mobility lower energy level in the canduction band
SEMICONDUCTOR MICROWAVE DEVICES 243

that is partially or fully filled get transferred to another low mobility,

high energy level that is empty when the applied electric field or the
applied voltage for a given semiconductor slice thickness exceeds a
certain threshold value. This electric field is 33 KVIcm for Gallium
arsenide. Since, we are referring to transportation of electrons, it should
be an N-type semiconductor. So, this figure is for N-type GaAs. Other
semiconductor materials which exhibit this effect and which have been
exploited for fabrication of practical Gunn devices include Indium
Phosphide (InP), Cadmium Telluride (CdTe), Gallium Arsenide Phos-
phide (GaAsP),Indium Arsenide (InAs) and so on. So for the transferred
electron effect to manifest itself in a bulk semiconductor material, the
material should possess some specific band structure. Firstly, the
semiconductor should have two energy sub-levels in the conduction
band, one partially or fully filled and the other empty. Secondly, the
lower and upper energy levels involved in the transfer process should
have a forbidden gap that is much smaller than the energy band gap of
the semiconductor. Otherwise, the complete crystal structure of the
device would probably get destroyed before any transfer of electrons
could take place. This gap is typically 0.36 eV i n GaAs while its
semiconductor band gap is 1.43 eV. The situation is depicted in Fig.
6.28. Thirdly, the electron mobility in the lower level should be much
higher than the electron mobility in the upper level. Also the effective
mass of electrons in the lower level should be smaller than the effective
mass of electrons in the upper level.

D
Wave Number Wave Number

Wave Number Wave Number

(c) (dl
Fig. 6.28
244 MICROWAVES AND RADAR
The transferred electron effect lead? a what is known as Negative
Differential Mobility which further causes the current density to
decrease for increasing electric field after it exceeds the threshold. Fig.
6.29 shows Current Density (J)versus Electric Field (E)characteristics
of a Gunn diode. The curve can be explailied as follows. When the

>
Eb Electric Field (E)
Fig.6.29
applied electric field E is less than the required threshold electric field
ET, the electrons in the lower level move towards the positive end and
contribute to current. Increasing field yields increasing current. This is
the case of a normal positive resistance. This situation is depicted in
Fig. 6.28(b) where the lower energy, higher mobility level in the con-
duction band is shown filled and the higher energy, lower mobility upper
level is shown empty. When the field equals ET, the electron transport
mechanism comes into play. The electrons move to a higher level having
a much lower mobility with the result that current decreases. This is
shown in Fig. 6.28(c).Further increase in electric field leads to further
reduction in current as more and more electrons are transferred to the
upper level. Finally, the applied electric field reaches a value where all
electrons are transported to upper level. Eventually, applied electric
field is strong enough to remove all electrons from the lower mobility
upper level as shown in Fig. 6.28(d) after which the device again
exhibits positive resistance. The device exhibits negative differential
mobility or resistance in the region between the electric field values
Ea and Eb provided pi. Ea> p2. Eb where 'pi' and ' p ~are' the mobilities
of the lower and upper levels respectively.
OperationalModes
The two main operational modes of Gunn diodes include :
1. Gunn Mode
2. Limited Space Charge Accumulation (LSA)Mode
The two modes are briefly described below.
Gunn Mode :An excess charge distribution in a material exhibit-
ing a positive resistance or conductance characteristics will always
disperse due to mutual repulsion forces as per the expression :
P = PO Exp (- t/Td)
where Td = Dielectric relaxation constant = E / O
SEMICONDUCTOR MICROWAVE DEVICES 245

E = Dielectric constant, p= Change density and o=Conductivity

This decay of excess charge is known as dielectric relaxation effect.
For the case where conductivity is negative, the dielectric relaxation
phenomenon becomes a dielectric growth phenomenon i.e. any small
perturbation in charge density grows instead of decaying. Let us see
how it applies to a Gunn diode when i t is biased in the negative
differential conductivity region. A space charge domain nucleated a t the
cathode travels towards the anode under the influence of the electric
field. As it travels, it grows up exponentially due to the device being
biased in the negative differential conductivity region. As a result, the
electric field in the region outside the domain falls below the threshold.
Formation of domain near the cathode can be explained as follows. The
doping concentration in the semiconductoris never absolutely uniform.
It is quite probable that some part near the cathode has less concentra-
tion. This region would therefore have less free electrons and hence less
conductivity or higher resistance. This means that for a given applied
voltage, this part has a higher mobility and therefore a higher prob-
ability of being the first to attain negative differential mobility. Nega-
tive differential mobility leads to bunching of electrons as those which
are behind catch up with the slower ones in the domain and the those
which are ahead move faster and go further away. This leads to
formation of a domain. Coming back to the domain growing up more
and more as it moves towards the anode, this grown up more mature
domain finally disappears at the anode. At this moment, when the
domain is disappearing, the electric field again rises above the threshold
and another domain is nucleated at the cathode. Each time the domain

Fig. 6.30
246 MICROWAVES AND RADAR

disappears at the anode, it gives rise to a current pulse in the external

circuit. The movement of the domain through the device length as a
function of time is shown in Fig. 6.30. The frequency of oscillations in
this mode is given by 'Vs/L' where Vs is the saturated drift velocity and
L is the device length to be traversed by the domain. V, for N-type
GaAs is about lo5 cmls.
Also, the domain must have enough time to mature or fully grow
before it makes an exit at the anode. That is, Td should be much smaller
than the time period of oscillations.
Or ( ~ / n ~v)
e .<< L/Vs which gives no L 2 2.5 x 1 0 ' ~ / c r n ~
In practice, no. L 2 3 x 1 0 ~ ~ / c m '
It should be noted that the Gunn oscillator does not require a
resonant circuit to operate. It could oscillate in a purely resistive circuit
as long as series equivalent load resistance is not too large.
Two variations of the Gunn mode are the Quenched Gunn Mode
and Delayed Gunn Mode. These are obtained by enclosing the Gunn
diode in a resonant cavity. In the Quenched Gunn Mode, the RF voltage
swing is large enough to extinguish a domain in transit itself at the
voltage minimum of the RF cycle as shown in Fig. 6.31(a). In this mode,
the domain never transits fully across the device length with the result
that the operational frequency is higher than the frequencyin the Gunn
mode. In the Delayed Mode, as shown in Fig. 6.31 ( b ) ,RF swing is large

ET=Thresholdfidd
Tc Tt ES=Sustaining field
T, = Transit time

t
-

Es ---------------- ------ ------

Tt t-
(b)
Fig. 6.31
SEMICONDUCTOR MICROWAVE DEVICES 247

enough to keep the field below the threshold during the negative swing
after a domain is exited. Formation of a new domain is delayed with the
result that operational frequency in this mode is lower than the Gunn
mode frequency. In both these modes, the operational frequency is
decided by the resonant cavity and not the transit time. Practical Gunn
oscillators use either of these two modes as they allow control over
frequency, give improved efficiency and a reduced carrier noise.
LSA M o d e :The DC to RF conversion efficiency in Gunn modes
is in the range of 1%to 5%.One of the reasons for this low efficiency is
that in the domain mode, when the domain is propagating, most of the
device is below the threshold field. The major part of the device thus
acts like a parasitic lossy resistance. If the domain formation could
somehow be prevented, the entire region could be made to participate
in generation of power and as a result, efficiency would improve.
In the LSA mode, RF has a large swing and a time period that is
small as compared to dielectric growth time. The peak to peak
amplitude of RF is such that for most of the time it is in negative
conductivity region and it swings below the threshold field for small
part of the RF cycle as shown in Fig. 6.32. The amount of space charge
that can accumulate during the time the device is in negative conduc-
tivityregionislimited. While the fieldis below the threshold, the limited
space charge gets dispersed due to relaxation effect. Since LSA mode
does not depend for its operation on the transit time effects, the device
can have a long active region with the result that very large pulse
powers of the order of hundreds of watts can be achieved from a single
large chip.

Fig. 6.32
The necessary conditions to be satisfied for the LSA mode are the
following :
1. R F cycle time period must be less than the growth time
constant so as to allow only a limited accumulation of space
charge.
2. Decay time constant must be much smaller than the RF cycle
time period.
248 MICROWAVES AND RADAR

The two conditions are satisfied if 2 x lo4 5 no / fo 22 x lo5/

cm3 /Hz.
6.11. Transit Time Devices
The family of transit time microwave devices depend for their
operation on two mechanisms namely the generation of charge carriers
and the movement of these charge carriers through a drift space withm
the seh;icorductor. Different devices belonging to this family can be
distitguished from one another on the basis of the mechanism used for
generation of charge carriers. The prominent members of this family
inclide the IMPATT and TRAPATT diodes, BARRITT diode, TUN-
NE'I'T diode and so on. The IMPATT (Impact Ionisation Avlanche
Transit Time! diode, as the name suggests, utilises impact ionisation
for carner generation. TRAPATT (Trapped Plasma Avlanche Triggered
Trhnsit cyme) diode is derived from the Impatt with some modifications
in the doping profiles so as to achieve higher pulsed microwave powers
a t trattx efficiency values. The BARRITT (Barrier Injection Transit
Time) diode employs carrier injection across a forward biased barrier
junction while in TUNXETT (Tunnel Injection Transit Time) diodes,
the carrier generation is by tunneling phenomenon. Other less common
devices are the MITATT (Mixed Tunnel Avlanche Transit Time),
QWITT (Quantum Well Injection Transit Time) and DOVETT (Double
Velocity Transit Time) devices.
Impatts and Trapatts are relatively noisier but are capable of high
CW and pulsed powers at reasonably good efficiencies with Trapatt
having an edge in terms of pulse operational efficiency. Tunnetts and
Barritts on the other hand are low noise devices but are capable of
generatinglower power levels only. Tunnetts have an added advantage
of operating a t much higher frequencies, in the region of tera hertz.
Some of the more popular devices of this family are described in detail
in the following paragraphs.
6.12. IZMPATT Diode
The IMPATT diode depends for its operation on the generation of
charge carriers by impact ionisation in a certain small region around a
reverse biased PN junction in a semiconductor material and the sub-
sequent transit of these carriers through a drift region within the
semiconductor. Silicon and Gallium Arsenide are the semiconductor
materials favoured for microwave IMPATT diodes. The device exhibits
a dynamic negative resistance across its terminals which is exploited
for its use in amplifiers and oscillators in the microwave and millimeter
wave region.
The IMPATT device was originally proposed by W.T. Read in 1958
to be followed by fabrication of first operational device by Johnston and
his associates in 1965. The IMPATT diode utilises the delay inherent
to the phenomenon of carrier generation by impact ionisation leading
to an avlanche and also the transit time delay to achieve the phase
difference of 180" between the input R F excitation voltage and the
terminal current pulse.
SEMICONDUCTOR MICROWAVE DEVICES 249

Fig. 6.33 sllows two possible device structures originally proposed

by Read for fabricatmg an IMPATT diode. in both types, there is a PN
junction and the avlanche action takes place in a very small region
around the reverse biased junction. Also, there is a single drift space
through whicl~the avlanche multiplied charge carriers have to drift to

Fig. 6.33
reach the device cathode (Fig. 6.33a) or anode (Fig. 6.333). The principle
of operation, which is nothing but the mechanisms of carrier generation
and avlanche multiplication and their movement through the drift
space to achieve terminal negative resistance, can be explained as
follows. When the device is sufficiently back biased so as to produce a
reverse electric field of hundreds of kilovolts per cm (typically 400 KV
/ cm),the minority carriers crossing the junction acquire so much energy
that they are able to knock out more electrons and holes from the crystal
structure, the process being called Impact Ionisation. The process of
impact ionisation leads to multiplication of charge carriers in what is
known as an avlanche condition. The IMPATT diode is so fabricated as
to be able to withstand such a high voltage gradient across the junction.
In the actual operation, the device is back biased so as to be at the
threshold of an avlanche to develop. Now, if an RF voltage exists across
the junction, as shown in Fig. 6.34 ( a ) ,then the avlanche would set in
a t the positive peak of this RF voltage which would lead to generation
of a high peak value current pulse at the junction as shown in Fig. 6.34
(b). It can be seen from the figure that the current pulse peak occurs
when the RF voltage is passing through zero and going negative rather
than occurring at the peak cf the RF voltage. This is due to the fact that
the avlanche process which is the process of current multiplication is
not instantaneous. Now, the current pulse at the junction lags the RF
voltage by 90'. The current pulse then drift through the drift region and
reaches the cathode (assuming the structure of Fig. 6.33 a). The thick-
ness of the drift region and the drift velocity decide the transit time.
And therefore the additional phase delay is introduced. The device is so
made that an additional phase delay of 90' is introduced due to transit
time effect. In fact, if 'T' is the transit time and I = l o Cos wt is the current
pulse at the junction, then the current pulse flowing in the external
circuit would be given by :
IeXt=In [(Sinm/2)/(w2/2)] Cos (at - ot/2)
250 MICROWAVES AND RADAR

If = x, then 'Iex$lags T by 90". This gives an overall phase shift

of 180" between the current flowing in the external circuit and the RF
voltage. And hence a terminal negative resistance. It can be visualised
here that the transit time that has to be equal to a certain fraction of
the RF cycle time period in order to achieve an overall phase difference
of 180" depends upon device thickness. For optimum DC to RF conver-
sion efficiency in an IMPATT oscillator, the phase angle corresponding
to the transit time should be equal to '0.74 n' radians. The efficiency
then is given by
q (IMPATI') = (2.27/x)[Vrf/Vdc]

Voltaae
RF t

Avlonche
Current

Fig. 6.34

Profile

Avlonche
Reg~on
I

Electric ~ i e l d p

Drift Region
i
;
X
Fig. 6.35
Fig. 6.35 shows the doping profiles and electric field distribution
of the Read IMPATI'.
SEMICONDUCTOR MICROWAVE DEVICES 25 1

The IMPATT diode structure discussed so far is the one originaily

- -
proposed by Read. The state-of-the-art commercial devices however are
usually fabricated with a single drift or double drift profiles. The doping
profiles and the electric field distributions of the two types are respec-
tively shown in Figs. 6.36 and 6.37.
If the device area is held constant, the double drift profile offers
the advantage of lower device capacitance and consequently higher
frequency of operation as compared to single drift designs as the double
drift design has greater inter-electrode spacing. Alternatively, for a
given device capacihnce, double drift device can be fabricated with a
greater junction area which in turn would offer a greater power han-
dling capability.

Fig. 6.36

Fig. 6.37
Applications :IMPATT diodes are used as microwave oscillators
and amplifiers over a frequency range extending well into millimeter
wave range. For single drift versions, CW output powers may vary from
20 W a t a few GHz to about 50 mW at 200 GHz. Efficiency varies from
10%to 20% up to a few tens of GHz reducing to 1% a t 200 GHz. Pulsed
252 MICROWAVES AND RADAR

output powers are still higher. Double drift versions produce greater
powers at higher efficiencies. A double drift silicon IMPATT from
Hughes produces a minimum of 1W CW power in the frequency band
of 58-62 GHz. Another double drift GaAs IMPATT from Raytheon offers
same specification at an efficiency of 10%.A silicon double drift IMPATT
from Hughes generates a pulse output power exceeding 20W in a
frequency band of 92-96 GHz.
The major problem associated with IMPATT devices is the noise.
Avlanche inherently is a noisy process. A high operating current adds
shot noise too. Nevertheless, they are in use due to their capability to
generate high powers at reasonably good efficiencies.
6.13. TRAPATT Diode
TRAPATT (Trapped Plasma Avlanche Triggered Transit Time)
diode is similar to IMPATT diode in structure and is usually of N+P+
or P+ NN+ type (Fig. 6.38). The device is back biased into avlanche like

Fig.6.38
the IMPATT. The difference between the two lies in the mechanisms of
avlanche initiation and carrier drift. A TRAPATT in operation is placed
in a high "resonant cavity and back biased to avlanche threshold. When
the RF oscillations begin, they build up extremely rapidly due to the
resonant structure thus taking the voltage across the diode to a value
much above the avlanche threshold. The result is generation of a
conducting plasma of holes and electrons which rapidly fills the entire
drift zone. As a consequence of this, the voltage across the diode falls
which does not allow the carriers in the plasma to escape as quickly as
they would have in case of IMPATT operation where there is a high
electric field in the drift region. The charge carriers are said to be
trapped and hence the name trapped plasma. The carriers are now able
to drift with a relatively much lower velocity due to the residual electric
field. As they reach the device electrode, they constitute a large current.
When the entire plasma is extracted, the voltage again rises above the
avlanche threshold and another cycle is initiated.
TRAPATT diode is essentially a pulsed device capable of operating
a t much larger pulse powers as compared to IMPATTs. The operating
frequencies are however much lower. Also, they are noisier than
IMPATTs. Pulse powers ofkilowatt level have been achieved at frequen-
cies around 10 GHz with efficiency approaching 25%.
6.14. BARRITT Diode
A BARRITT diode uses punch through injection and the transit
time mechanisms to build a two terminal semiconductor microwave
negative resistance device. The practical significance of a BARRITT
diode lies in the fact that it is easy to fabricate and it can be made using
SEMICONDUCTOR MICROWAVE DEVICES 253
silicon technology. It is a low noise device with its noise performance
superior to that of an IMPATT device due to absence of avlanche
mechanism. It is however less efficient than an IMPATT device and it
is for this reason that its use would mainly be limited to low power
applications such as local oscillator for radars etc.
Fig. 6.39 shows the basic device structure of a BARRITT diode. It
can be assumed to be pair of two abrupt PN junctions connected back
to back. It consists of a lightly doped N-region sandwiched between two
heavily doped P-regions. Practical BARRITT diodes use schottky bar-
rier junctions. When the applied voltage is say zero, the PN junction at
the left (X = 0 ) is slightly forward biased while that at right (x = L) is
slightly reverse biased. The carriers (holes) injected a t the left side
junction from P+ side will have practically no chance of diffusing
through to the P+ region on the right side as would be clear from the
electric field plot for V = 0. The current that can flow under the cir-
cumstances is the reverse biased junction's leakage current. As the

voltage is increased, the depletion width ofleft junction decreases while

that of right junction increases as is shown in Fig. 6.39. 'The current
even now also would be restricted to reverse biased current. As the
applied voltage is increased further, a stage comes where the depletion
region of the right side junction reaches through t o the forward biased
junction as shown. I t is now when all the injected carriers (holes),which
have also increased substantially due to increased forward bias are
swept across the device from left to right under the influence of electric
field of depletion region. This leads to rapid increase in diode current.
The injected electron current moving from right t o left can be ignored
because of minority carrier suppression in Pf region. As the voltage is
increased still further, space charge limiting sets in as the injected hole
density approaches background doping density. The current then in-
creases very slowly. The V - I characteristics are shown in Fig. 6.40.
254 MICROWAVES AND RADAR

Fig. 6.40
The characteristics have three distinct regions defined by
1. V < Vpt, I = I(leakage)
2. V=Vpt, I = Exp ( b V )
3. V > Vpt,I increases slowly
If V is increased still further, avlanche breakdown of device can
occur. The punch through voltage can be expressed by :
Vpt = [e.N&E] L~
where L = N - region width
N d = Doping concentration
The electric field E (x) as a function of 'x' can be expressed by :
E(x) = [e. N ~ Ex ] + [(Va - Vpt)/L I
which gives Emax = e.Nd . L I E
BARRITT diode oscillates a t a frequency that gives a transit angle
of '3d2' which gives the oscillation frequency as :
fo = 3VS/4L

6.15. TUNNETT Diode

The TUNNETT (Tunneling and Transit Time) diode, acother
device belonging to the family of transit time devices, depends for its
operation on the tunneling mechanism for generation of charge carriers
and transit time for transport of these carriers. Since the tunneling
mechanism occurs a t a relatively higher electric field (typically > 1000
KVIcm)in contrast to avlanche process that is prevalent atlower electric
field (typically < 500 KV/cm),it is necessary to design the doping profile
of TUNNETT diodes suitable for generating high electric field. Struc-
turally, it is similar to the IMPATT diode so that the processing
technology used and perfected for fabricating IMPATT diodes can be
used for fabrication of TUNNETT diodes. Due to the fact that tunneling
is a quiet process as compared to avlanche mechanism, TUNNETT
diodes offer much better noise performance than the IMPATT diodes.
SEMICONDUCTOR MICROWAVE DEVICES 255

Also due to very short transit times involved in its operation, these are
capable of operating at higher frequencies, comfortably well beyond 100
GHz. However, TUNNETTs are low power devices and certainly do not
match the IMPATTs in this respect. Due to these above mentioned
properties and limitations, TUNNETT diodes are found to be very
useful as a low noise low power local oscillators and often a better
replacement for Gunn diode local oscillators at frequencies greater than
the upper limit of Gunn &odes.
The optimum DC to R F conversion efficiency in a TUNNETT diode
oscillator occurs when the phase angle corresponding to the transit time
is '1.5 n' radians. For this condition, the conversion efficiency turns out
to be equal t o :
(TUNNETT)= (2/37c) [Vrf/Vdc]
The above expression for conversion efficiency shows that the
IMPATT oscillator would have more than three times better efficiency
than a TUNNETT diode oscillator. This explains why TUNNETT diode
oscillators are suitable for low power applications only.
6.16. MITATT Diode
MITATT (MixedTunneling and Avlanche Transit Time) diode uses
both tunneling and avlanche mechanisms for generation of charge
carriers and therefore has the noise performance and the power
capability that is between those of a TUNNETT and an IMPATT.
6.17. Microwave Field Effect Transistors
Basic operational mechanism of a microwave FET is no different
from that of a conventional junction FET that we are familiar with. The
operation is based on the variation in the drain source channel current
as a function of applied gate source voltage. The width of depletion
region and the conducting channel are controlled by the longitudmal
electric field resulting from applied gate source voltage for a given drain
source voltage. A more negative Vgs leads to a wider depletion region
and consequently a narrower channel and a reduced Ids (assuming an
N-channel device).Ids is maximum for Vgs = 0 for a given V h and the
magnitude of Vgsthat widens that depletion region to the extent of
completely blocking the channel is called the 'Pinch-Off Voltage'. At
Pinch-Off, the drain current is zero.
Fig. 6.41 shows the I-V characteristics of a microwave FET. The
characteristics can be divided into three distinct regions. The first
marked '1' is the Linear Region where Ids is proportional to VdsThis is
the region where carrier velocity is proportional to the electric field (Fig.
6.42). The second region marked '2' is the Saturation Region where
Ids remains constant irrespective of change in Vds. Infact, this is the
region where the charge carriers are moving with saturated velocity. A
change in Vds or in other words the electric field, under whose influence
the charge carriers move, in this case does not produce change in carrier
velocity. The third region marked '3' is the Breakdown.Region where
Ids increases very rapidly with even a slight increases in Vds.
256 MICROWAVES AND RADAR

Vd s

Fig. 6.41

Fig. 6.42
Types Of Microwave FETs
Coming to different types cfm~crowaveFETs, there are two broad
categories namely the Metal Semiconductor Field Effect Transistor
(MESFET) and Heterostructure Field Effect Transistor (HFET) also
called High Electron Mobility Transistor (HEMT).
MESFET : MESFET is a junction FET employing a schottky
barrier junction for the gate and made from Gallium Arsenide (GaAs)
or Silicon (Si) or Indium Phosphide (In P) with Gallium Arsenide being
the most commonly used semiconductor material. Silicon MESFETs
have a lower operating frequency while InP MESFETs suffer from the
disadvantage of lower gain. GaAs MESFETs are capable of operating
well into millimeter wave region.
Fig. 6.43 shows the cross-sectional structure of a GaAs MESFET.
Drain and Source electrodes form ohmic contacts with the active layer.
Gate electrode forms a Schottky barrier. Gate length and width are very
important parameters. While gate length determines maximum opera-
tional frequency, noise performance and gain, gate width decides RF
power capability. Decreasing the former increases the frequency of
operation, reduces noise and improves gain. GaAs MESFETs capable
of operating at millimeter wave frequencies have a gate length of about
0.25 micron. Gate widthis usually increased by having multiple fingers
interconnectedvia air bridges. Overall width may be in the range of400
to 1000 microns in case of power MESFETs.
SEMICONDUCTOR MICROWAVE DEVICES

Source Gate Drain

1
Semi-insulatmg
GaAs Substrate

Fig. 6.43
HFET (or HEMT) : It is a conventional MESFET fabricated on
an AlGaAs / GaAs heterostructure. Another type of HEMT called
Pseudomorphic HEMT (PM HEMT) is based on an AlGaAs / InAIAs /
I n G a h heterojunction. Yet another type is InP based HEMT based on
InAlAsAnGaAs heterojunction and fabricated from InP substrate.
PM HEMTs and InP HEMTs outperform GaAs MESFETs in terms
of many important performance parameters like Noise Figure, Gain,
Noise Bandwidth, Gain bandwidth product with InP HEMTsoffering
the most superior performance standards of the three types mentioned
above. InP and PM HEMTs have been developed offering Noise Figure
as small as 1.6 dB at a frequency approaching 100 GHz.
Cross-sectional structures of conventional HEMT, PM HEMT and
InP based HEMT are shown in Figs. 6.44(a),( b )and ( c ) respectively.
The diagrams are self explanatory.
Equivalent Circuit Of a Microwave FET
Fig. 6.45 shows the equivalent circuit of a microwave FET. The
diagram has been intentionally drawn as a combination of the device
structure and the equivalent circuit to make it a bit more illustrative
Rg,Rs and R d are gate, source and drain electrode resistances. Ri is the
intrinsic channel resistance Cgs,Cgd and Cds are the gate source, gate
drain and drain source capacitances. R d s is the output resistance and
g, is the transconductance. Ids is the drain source current. Lg& and
L d are gate, source and drain electrode inductances.
Important Parameters
The important parameters characterising microwave FETs in-
clude :
1. Unity gain cut-off frequency, f i
2. Maximum operational frequency, fmax
3. Maximum available gain at a given frequency, GmW
4. Unilateral power gain (Power gain with no reverse transmis-
sion), Gu
5. Minimum noise figure, fmin
The relevant expressions for computing these parameters are :
- - - -

where us = Saturated velocity

258 .MTCROWAVES AiiD RADAR

L = GAe length

Rg+Ri+Rs
where rl =
Rds

It can be seen from the above expressions that f~ plays a very

important role in determining the noise performance of the device.f~ is
related to transit time under the gate. Designing a device with mini-
mum transit time results not only in a high cut-off frequency but also
in a good noise performance at a given frequency. Constant Kfindicates
quality of the material usedin fabricating the FET and is approximately
2.5 for a MESFET.

N-type AlGoAs N-type AI GoAs

AlGaAs Spocer
I GaAs Buffer 1 In GoAs

Semi insu!at~ng
GaAs Substrate

I Semi-~nsulatmg
GaAs Substrate

1 InAlAs Spacer
In GaAs
I
In AlGaAs Buffer

InP Substrate

(c )
-- 1
Fig. 6.44
SEMICONDUCTOR MICROWAVE DEVICES

I Substrate
I
Fig. 6.45
6.18. Maser
A MASER (acronym for Microwave Amplification by Stimulated
Emission of Radiation) is not really a semiconductor microwave device,
not even close to the semiconductor devices that we have discussed in
the chapter so far in terms of structure o r operational principle. It is a
device that provides extremely low noise amplification of microwave
signals by a quantum mechanical process. This new idea came from
Townes and his colleagues in 1954. It is the utility of maser as a low
noise amplifier at microwave frequencies for which host of semiconduc-
tor microwave devices were developed, that it has been chosen to be
discussed in the present chapter. Townes extended his newly discovered
idea to generate and amplify optical frequencies i.e. light in another
device called LASER (acronym for Light Amplification by Stimulated
Emission of Radiation), which we all know later turned out to be a
revolutionary device. This work was considered so significant that it
fetched him Nobel prize in Physics jointly with two Russian scientists
Basov and Prokhorov in 1964 for their work in Maser and Laser
principles. While maser fundamentals and applications are discussed
in this section, a brief on lasers would follow immediately thereafter.
Operational Principle
It is an established fact that electrons in any material, due to their
motion around their respective nuclii in different orbits, occupy certain
discrete energy levels. In normal conditions of temperature and in the
absence of any other energyinput, these would occupy only lower energy
levels. If the material is supplied with a quantum of energy which is
such that AE = hfwhere AE is the energy difference between an occupied
 260 MICROWAVES AND RADAR

lower level and another possible or allowed (though empty) upper level,
then electrons in the occupied level will absorb this energy quantum
and go to the higher energy level. The electrons are then said to be in
the excited state. After some time, they would probably fall back to the
lower level re-emitting energy of the same frequency as that of the pump
source. Now assume that the material has allowed energy levels like
the one shown in Fig. 6.46. the energy difference between the upper
level and the intermediate level corresponds to the desired microwave
frequency to be amplified. If such a material is kept in a resonant
structure and pumped with microwave energy such that the energy
quantum supplied (= hfl equals the energy difference (E2 - EO). The
electrons will get pumped from the lowermost level, also referred to as
ground state, to the upper level. Now if the resonance frequency of the
resonant structure is such that it equals the frequency corresponding
to (E2 - E l ) , then any microwave input at a frequency corresponding

, ' :: ;
Emission

II Absorption

Fig.6.46
to an energy difference (E2 -El) will get amplified. Infact, the input
microwave energy stimulates the excited electrons at the Ez-level to fall
to El-level and the process is helped by the resonant structure.
Electrons, during their transition from E2-level to El-level give their
energy to the input which in turn gets amplified.
A device like this would be an extremely low noise device. There
is no resistance involved in the amplification process which means that
thermal noise would be absent and also there is no moving stream of
electrons to add shot noise. In addition, if the device is cooled to
cryogenic temperatures, it reduces the noise even further.
Thus for a maser device to function, what is needed is to have an
appropriate material with desired energy level structure, a suitable
pump source and desired resonant structure. The substance used
originally in the maser device was the ammonia gas. The maser using
ammonia gas had the disadvantage of operating over a narrow band of
frequencies of about 3 kHz at 24 GHz. The first practically useful maser
was built with a solidstate material called Ruby which is a crystalline
form of silica (&03) with a slight doping of chromium. Ruby had all
the desirable characteristics required for a maser device. It had suitable
energy band structure and was paramagnetic. The slight magnetic
property permitted some tunability. The energy levels of Ruby could be
altered by changing the strength of a strong magnetic field of the order
of 4000 A/m.
SEMICONDUCTOR MICROWAVE DEVICES 261

Operational Ruby maser is invariably cooled to liquid helium

temperature as it is observed that Ruby has significant population of
electrons at the upper level at room temperatwe. Cooling of Ruby
corrects that situation and depopulates the upper level so that large
number of electrons can be pumpedfrom groundlevel to the dpper level.
In addition, cooling significantly improves noise performance. Fig. 6.47
shows the schematic diagram of a cryogenically cooled Ruby maser.
Ruby maser too has a narrow bandwidth (typically 1.5 MHz at 1.5GHz)
but is much better than that of ammonia maser. Bandwidth could be
increased by using a traveling wave structure like the one discussed
earlier in case of traveling wave parametric amplifier.

from antenna- p t o mixer

Waveguide Liquid
(for Pump Nitrogen at 77K
input )
Circulator

Liquid Helium
at 4.2K

Fig. 6.47
The biggest advantage of a maser (Ruby maser) is its excellent
noise performance. Noise figure of better than 0.3 dB is common.
Bandwidths of about 25 MHz at 1.5 GHz are achievable with TW
masers. The main disadvantage is that it is a low level amplifier that
would saturate with a few microwatts of power. Parametric amplifiers
have replaced masers in most applications as their noise performance
approaches that of masers. Masers are used where the received signals-
are very weak like those encountered in Radio Telescopes, Communica-
tion Receivers used in conjunction with space borne probes and so on.
6.19. Laser
The word LASER is an acronym for Light Amplification by Stimu-
lated Emission of Radiation. The principle of o~erationof a laser is an
extension of the concept underlying the principle of operation of a maser
device. Both the devices rely for their operation on the concepts of
Stimulated Emission OfRadiation and Population Inversion, the latter
being a pre-requisite to achieve the former. The basic components of a
laser device too are an appropriate material called the Laser Material
that possesses the right energy level structure, a Pumping Source
capable of producing population inversion and a resonant structure
262 MICROWAVES AND RADAR

usually called Resonant Cavity to strengthen the process of stimulated

emission and help build the stimulated emission by using feedback.
The lasers are either three level or four level systems depending
upon the number of energy levels involved in the operation. The Ruby
maser was a three level device and so is the Rubv laser which can take
the credit of being the first functional laser de"vice demonstrated by
Maiman in 1960. In a three level laser, the energy levels involved are
the gound level which is also the lower laser level, the upper laser level
and a s ~ lhigher
i energy level called the highly excited level (Fig. 6.48).
The p,npi.~g Eource excites the atoms in the ground level to the
upprrmcist highly excited level. The excited atoms quickly fall to the
upper lasw level also called the metastable level. The laser action takes
place oehsen this metastable Iwel and the lower laser level which is
the grou-id level in this case. The frequency of emitted radiation
corAespo,.cisto the energy difference between the two laser levels. The
putlipiqg pocess creates population inversion between the two laser
levels v-Lich is necessary for the stimulated act~onto occur. It may also
ba mnncionsd here that a much longer lifetime for the metastable state
as compared to that of the highly excited level also helps in building the
population inversion. A three level arrangement like this however has
a disadvantage that due to the ground level being the lower laser level,
it becomes necessary to lift majority of atoms from the ground level to
the upper level to be able to create population inversion as the ground
level is highly populated. This means a high lasing threshold energy.
Ruby laser is one of the most important types of three level laser devices.
Highly Excited Level
T.

$11-
.-
V
11 Laser Transition

L Ground State

Fig. 6.48
In a four level laser, the laser transition is between the metastable
level and a lower lev4 which is not the ground level as shown in Fig.
6.49. The basic difference between a three level and a four level system
is that it is much easier to create poplation inversion in a four level
system in terms of pump energy to be supplied to the system due to the
fact that the lower lasing level in a four level system is not heavily
populated like the ground level. As a result, it is necessary to lift only
a small quantity of atoms to the upper stat2 for the lasing action to
become possible. This means that these lasers would have alower lasing
threshold. In addition, if the lower laser level has a life time that is
much shorter than that of the upper laser level, it helps in depopulating
the lower laser level which is equivalent to populating the upper laser
SEMICONDUCTOR MICROWAVE DEVICES 263
level as far as creating population inversion is concerned. Such lasers
are capable of sustaining population inversion on a continuous basis
and thus provide a continuous (CW) laser output.

Upper Laser Level

roundState
Fig. 6.49
It is also worth while mentioning here that in most real lasers,
there may not be a single upper laser level and a single lower laser level.
There may be more than one closely spaced upper levels or lower levels
as shown in Fig. 6.50. This leads to the possibility of having laser action
on multiple transitions thus producing hfferent output wavelengths by
suitably designing the optics.
H~ghlyExc~tedLevel

}upper Laser Level

1/ 9-
rewo~) , Laser Level

J Ground State

Fig. 6.50
Based on the type of laser material used, there is a large variety
of lasers. All forms of media i.e. solids, liquids and gasses hav.e been
exploited to demonstrate laser action. The more commonly used types
amongst them are the Chromium doped silica (Ruby), Neodymium
doped Yttrium Aluminium Garnet (Nd-YAG),Neodymium doped Glass
(Nd-Glass)in the category of solidstate laser materials, Carbon dioxide
and Helium-Neon in the gas laser category and Gallium Aluminium
Arsenide (GaAlAs), Indium Gallium Arsenide Phosphide (InGaAsP),
AlGaInP (Aluminium Gallium Indium Phosphide) in the semiconductor
laser category. While discussing laser materials, it is important to
mention here that the active media in many practical lasers would
contain more than one species of atoms or molecules. One of the species
would be an efficient absorber of pump or excitation energy. This species
264 MICROWAVES AND RADAR

would then transfer the acquired energy to the other species (lasing
species) to create population inversion and subsequent laser action.
The commonly employed pumping mechanisms include pumping
by optical means and pumping by electrical means. Solidstate lasers
are optically pumped either by using a flash lamp or by using an array
of diode lasers. The electrical pumping is either by electrical discharge
as is the case in carbon dioxide and He-Ne lasers or by electrical current
through the diode junction in semiconductor lasers.
Characteristics Of Laser Light
Laser light has some special properties that makes it stand apart
and well above other conventional light sources. These include
Coherence (both temporal and spatial), Monochromaticity and Direc-
tionality. Coherenceis the most important of all and the other properties
are in some way or the other the outcome of the coherence property of
the laser. These properties are all inherent in the process of stimulated
emission due to which all photons that are emitted by the stimulation
process would have the same frequency, phase and direction.
Coherence of a travelinglight wavefront implies that different light
waves in the wavefront are in phase with each other. The coherence can
be temporal or spatial. If the phase along a traveling wavefront a t a
particular instant of time in the direction of propagation is the same as
its phase at the same point after the wave has traveled a distance 'x'
irrespective of distance 'x' traveled, the wave is said to be ideally
coherent temporally. In real lasers, 'x' is not infinite and it is referred
to as the coherence length. The time taken by the light wave to travel
that distance (= x / c ) over which the phase is maintained is called the
coherence time. Spatial coherence refers to having the same phase in a
direction transverse to the direction of propagation. While temporal
coherence refers to coherence as a function of time, spatial coherence
on the other hand measures the area over which the wave is coherent.
A laser wavefront that is highly coherent spatially would have a very
narrow divergence. As mentioned above, coherenceis inherent to stimu-
lated emission process responsible for laser action. The emitted beam
would be coherent if different photons are emitted from the same
position with the same phase and also that they have same frequency
or wavelength. While temporal coherence relates to the degree of
monochromaticity of the emitted radiation, spatial coherence is deter-
mined by transverse mode discrimination property of the resonant
cavity. Monochromaticity refers to the single frequency (in the ideal
case) or extremely narrow bandwidth (in real lasers) nature of the laser
light. A carbon dioxide laser, for instance, emitting at 10.6 microns (=
12250 GHz) has a frequency spread of 60 MHz only which could be
reduced to even a few tens of kHz by using frequency stabilisation
techniques. The Directionality of the laser beam also follows from the
coherence of the stimulated emission process.
Applications
Lasers have a very wide spectrum of applications. Their military
applications include their use in Laser range finders, Laser designators,
Laser trackers and recently their use as a speed of light laser beam
SEMICONDUCTOR MICROWAVE DEVICES 265

weapon. They are used by industry for various kinds of materials'

processing applications like cutting welding, shaping and so on. The
commercial applications include their use in Laser printers, Laser
based optical character readers, Laser compact disc audio players,
CD-ROMs, Lasers in long distance fiberoptic communication etc. They
have a whole range of biomedical applications both for diagnostic and
therapeutic purposes. Their scientific and research related applications
include Laser holography, Laser spectroscopy, Laser interferometry
and so on.

SOLVED PROBLEMS
Problem 6.1 : A microwave signal source has a n output conduc-
tance of 0.0025 mhos matched to a load for maximum power transfer. A
negative resistance microwave device having a negative resistance of
-250 ohms is placed across the circuit as shown in Fig. 6.51. Prove that
the overall circuit is capable of providing amplification and that th.e
power gain is 25.

Fig. 6.51
Solution :With the negative resistance device not connected, the
power that would be transferred to the load is given by :
P1 = V L ~ ~ L
Since the source is matched to the load, g~ = 0.0025 mho
2
Therefore, P = V Z ? ~ L= V L = [~Z . /~~ L ~+
gs)] x gs

With the negative resistance device connected,

Igs
r 12

J
1.3
Load power, PZ = vL2
gL =
+gL - gL

2
p2
Therefore, - =
I s gL 4 g ~
P* (2gL - g ) z X ~
266 MICROWAVES AND RADAR

P2
For - > 1, 4 g >~g so that denominator is less than numerator
P1

g~ = gs = 0.0025 mho
I p2 I -
Therefore --
4(0.0025)~ - 4 x 625 x 10" = 25
I p1 1 ( 2 x 0.0025 - 0.004)' - lo-6

Problem 6.2 :What should bc the maximum negative differential

conductance of a microwave device if it is to bc used for amplifying a
microwcve signal where the source of microwave signal feeds a load of
500 R
Solution : Load resistance = 500 Q
1
Load conductance = -= 0.002 rnho = g~
500
4 g =~ 4 x 0.002 = 0.008 mho
The necessary condition for amplification isg < 4 g where
~ 'g' is the
magnitude of negative conductarlce.
That is, gmax=4 g= ~0.008 rnho

Problem 6.3 :A Barritt diode employs a 10 micron wide N-region.

If the saturated velocity of carriers is assumed to be lo5 m l s , idetermine
the operational frequency when the diode is used as a n oscillator.
Solution :The oscillation frequency, (fo) is given by :

where V s = Saturated velocity of carriers

L = Length of N-region (drift region)

Problem 6.4 : A MESFET has a gate length of I micron. What

should be the unity gain cut-off frequency for this device if the carriers
move with a saturated velocity of 107 c m 1 s under the gate region ?
1
Solution : f~ =-
27T zt
where f~ = unity gain cut - off frequency
SEMICONDUCTOR MICROWAVE DEVICES 267

tt = Transit time

where L = Gate length

V s = Saturated velocity
vs
Therefore, fr = -
2xL

Problem 6.5 :A Gunn diode oscillator is oscillating at 10 GHz

while operuting i n the Gunn mode (Transit time mode). If the saturated
carrier velocity in the semiconductor is l o 7 c m l s , determine the length
of active layer.
Solution :f = 10 GHz = 10'' Hz
V s = lo7 c m / ~= lo5 m/s
In the transit time mode, i Z'L' is the length of active layer,
f~ = vs

Problem 6.6 : In Problem 6.5, what should be the doping

concentration for the Gunn mode to prevail satisfactorily given the
follow-ing :
Dielectric strength, E., = 13
Mobility, p = 100 crnZ/v-s

e :? 1.6 x 10-19 C
Solution : The necessary condition for Gunn mode to prevail is
Td << T
where d = Dielectric relaxation time
E L
NOW, -< < -
now vs
where no = Doping concentration
268 MICROWAVES AND RADAR

Problem 6.7 :For the Gunn diode of Problem - 6.6, prove that the
diode is operating i n LSA mode if it is oscillating at 4 0 GHz. Assume
mobility in the positive conductance region (p') to be 8000 cm2/v-s and
doping concentration, no = 10 15/cm 3 .

Solution : -= lo5 = 2.5 x 104/cm3/Hz

fo
40 x lo9
The necessary condition to be satisfied for the diode to operate in
LSA mode is given by :

The necessary condition is satisfied

Problem 6.8. Calculate the dielectric relaxation time, for G d s

given the following data for G d s .
~r = 13

Assume e = 1.6 x lo-'' C, EO = 8.85 x 10-l2 F / m

Solution : Dielectric relaxation time (Td) is expressed by :
E EOEr
Td=--=--
nCle nw
SEMICONDUCTOR MICROWAVE DEVICES

Problem 6.9 : A Gun~zdevice operates i n the transit time mode at

20 GHz. If it is fabricated from Gallium Arsenide, find length of device.
Assume V . = 107 crn/s, $ = 8000 cm2 / V-s , I.L = 150 crn2/ V-s,
n = 1 0 ~ ~cm3.
1
Solution : In the transit time mode,
f=-
Vs
L

Problem 6.10 : Fig. 6.52 shows Velocity versus Electric field curve
for a Gunn diode. Determine the mobility values for this diode i n the
positive and negative conductance regions.

Fig. 6.52
Solution :

Problem 6.11 : A Barritt diode has a 10 Frn wide active layer. If

the velocity of carriers is 8 x lo6 cmls, determine the electric field at a
point half c.laydown the active layer when the applied voltage equals the
270 MICROWAVES AND RADAR

punch through voltage. Also determine punch through voltage. Assunie

n = l o t 5 / cm3 and ~r = 12.5.
Solution :The electric fieldE(x a; a function of length ( x ) is given
by :

where Va = Applied volt'lge

V ~ =T Pmxh &rough voltage
L = Length of active layer
L
For Va = V ~ and
T x =-
2

It can also be computed as a product of electric field at an applied

voltage equal to (VPT) and the lengtli of the active layer.
Therefore, Vpp = 6.2 x lo4 :: 1c3
= 72 Volts.

Prcblem 6.12 : A microwave bipolar transistor has a n @)

specification of 10 GHz. What should be the approximate value for its
) an operating frequency of (a) 2 GHz (6) 10 GHz.
( ~ F Eat

Solution : h w= b)
\ 1
where f = operating frequency
10
( a ) f = 2 GHz gives ~ F =E -= 5
2
(b)f =
10 = 1.
10 GHz gives ~ F =E -
10

Problem 6.13 : A semiconductor material has a relative pennit-

tivity of 15 and a conductivity of 133 m h o / m . Determine its dielectric
relaxation time constant. If the doping concentration is 1015 1 crn3,
determine the carrier mobility i n the material. Assume e = 1.6 x lrl' C,
€0 = 8.85 x 10-l2 F l m .

Solution :Dielectric relaxation time constant, Ta.=

SEMICONDUCTOR MICROWAVE DEVICES 271

where o = conductivity

= 10-l2 s = 1 ps
Now o = npe
where n = Doping concentration
p = Mobility
0
Therefore, p = -= 133 = 0.8312 rn2/v-s
ne 1015x lo6 x 1.6 x 10-l9-

Problem 6.14 :Fig. 6.53 shows the typical equivalent circuit of a

GcAs MESFET. Determine device$ ( f ~ jAlso
. determine i f this device can
be operated at a frequency of 40 GHz. Assume g , = 50 milli-mho.

Fig. 6.53
Solution : f i = - gm
2n: c g s
- 50 x
= 13.3 GHz.
2 x 3.14 x 0.6 x 10-l2 -
Also, maximum usable frequency, ha) is expressed by

where 73 = 2x: RgCgd

and rl =
Rg+Rs+Ri
Rds
272 . MICROWAVES AND RADAR

NOW rl= + + 2.5 = 0.018'75

400

13.3 x lo9
Therefore, fmax.=
2 40.01875 + 13.3 x lo9 x 282.6 x 10-l5
-- 13'3 logG 44.3 GHz.
- 0.3
As maximum usable frequency fmax = 44.3 GHz, operation at 40
GHz is therefore theoretically possible.

Problem 6.15 : A parametric amplifier operating i n upper

sideband up-converter mode is pumped by a 20 GHz pump frequency. If
the signal frequency is 2 GHz, determine the power gain i n dB. If the
parametric amplifier was operated as a non-degenerate lower sideband
up-converter, what would have been the powergain i n d B i n that case ?
Solution : If fl, f2 and f3 are the signal frequency, the pump
frequency and the frequency of the output signal, then according to
Manley-Rowerelations.
I
- P3 =I Power Gain = -
f3
I p1 I fl
when the parametric amplifier is operated as an USB up-converter,
f3 = fl + f2

Therefore, Power Gain = f1

- +f2 -
--2o 20 - 11 = 10.4 dB
+

fl 2
When the parametric amplifier is operated as a LSB up-converter,
f3 =f2 - f l
Therefore, Power Gain = -
f2-f1 =-
2 0 - 2 --9 = 9 . 5 4 d ~
fl 2

Problem 6.16 : Fig. 6.54 shows the simplified energy level

diagram o f a four level laser device. Determine the expected output laser
wavelength.
Solution : From the energy level diagram, the lower and upper
laser levels are characterised by energy levels of 0.25 eV and 1.5 eV
respectively with reference to the ground state. The energy difference
between the two levels is given by (1.5-0.25) = 1.25 eV.
Now, if 01 is the stimulated emission frequency, then
h f = 1.25 eV
SEMICONDUCTOR MICROWAVE DEVICES

: - Pump bond

! I T
Upper laser level

3 t ~ o w e laser rlevel
m
CY
o Ground level

Fig. 6.54
where h = Plancks constant = 6.63 x lo3* Joule-sec
hf = 1.25 eV

3 x los
Therefore, output laser wavelength, h = ------= lo4 m = 1 p
3 1014

6.20. Microwave Integrated Circuits (MICs)

The Integrated Circuits (ICs) developed for performing various
individual circuit functions such as Amplification, Oscillations, Mixing,
Frequency conversion, Phase shifting etc. and also a combination of
various functions to make a complete sub-system at microwave frequen-
cies are broadly known as Microwave Integrated Circuits (MICs).Based
on the nature of signals handled by them, they are categorised as Analog
ICs and Digital ICs. Based on the techniques used to fabricate these
ICs and their architecture, so to say, they are classified as Hybrid ICs
and Monolithic ICs. The hybrid ICs at microwave frequencies are
simply known an Microwave Integrated Circuits (MICs) as they were
the first to find a application in microwave systems. The second class
of ICs called Monolithic Integrated Circuits meant for use at microwave
frequencies are called Monolithic Microwave Integrated Circuits
(MMICs). Hybrid and monolithic microwave integrated circuits are
briefly described in the following paragraphs.
Hybrid And Monolithic Integrated Circuits
Hybrid MICs
Hybrid microwave integrated circuits (MICs) are a combination of
discrete active devices including even the individual monolithic circuits
274 MICROWAVES AND RADAR
and the passive components fabricated using thick film or thin film
technologies. The active devices like transistors, FETs or individual
nlonolithic ICs are bonded onto the passive conlponents and the inter-
connections made using thick filndthin film technology on a single
insulating substrate (Fig. 6.55). Depending upon the technology used
for fabricating passive components and the interconnections, the hybrid
ICs are either Thick film hybrids or Thin film hvbrids though it is not
unconmlon to find hybrid ICs that make use of both thick film and thin
film technologies.
In case of thick film hybrids, as mentioned above, passive com-
ponents such as resistors, capacitors and interconnections (which are

1 1
Fig. 6.55
basically conductors) are made in the thick film form and the active
devices in the form of bare chips are bonded onto the required locations.
Thin film hybrids on the other hand are those MICs in which resistors,
conductors, capacitors etc. are made by depositing resistive, conductive
or dielectric films of thickness ranging from few tens of nanometers to
a few microns on an insulating substrate. In case of thick film hybrids,
the film thickness is typically 10 to 50 microns.
Hybrid MICs have performance features almost comparable to
those of MMICs. In addition, they have considerable design flexibility.
When the requirement is in small numbers, they turn out to be much
cheaper than MMICs performing similar functions.
Monolithic Microwave Integrated Circuits (MMICs)
An MMIC is essentially a microwave IC where all components,
both passive and active, and their interconnections are made on the
SEMICONDUCTOR MICROWAVE DEVICES 275

same semiconductor substrate (Fig. 6.56). It may be mentioned here

that in case of an MMIC, substrate has to be semi-insulating to
accon~n~odate the required transn~issionlines and passive components.
The active components are made in isolated conducting regions of the
senliconductor material. For making interconnection lines, which are
almost invariably microstrip lines discussed at length in an earlier
chapter, and passive components, it is important that the substrate
forms a low loss dielectric. MMICs have the advantages of a smaller
size, higher reliability and a much lower cost per unit, provided that
the production numbers are large, and better reproducibility. Disad-
vantages include higher noise and cross talk, higher unit cost for
smaller numbers, inflexibility and so on. With technological advances,

Fig. 6.56
however, these disadvantages are being gradually overcome and
MMICs are replacing hybrid MICs in various applications.
Applications of MICs
MICs, particularly the MMICs, find extensive applications in
Satellite communications, Telecommunication systems, Smart
weapons, Phased array radars, Surveillance and countermeasure sys-
tems, Instrumentation, Supercomputers, Direct broadcast satellite TV,
Global Positioning Satellite (GPS),High Definition Television (HDTV),
Mobile telephone etc. Those applications where the required numbers
are relatively small such as defence related applications, the per unit
cost is large. In commercial applications such as Mobile telephone,
Direct Broadcast Satellite etc., where the consumption is in very large
numbers, the per unit cost of the MMIC is low.
276 M K .tOW.4VES AND M A R
Most of the MICs (both hybrid and MMICs) are analog in nahre.
MICs, for instance those developed for use in supercomputers, offer
digital functions. There are MICs which offer a blend of analog and
digital functions. While in MICs offering digi~alfunctions, the choice is
invariably the MMIC, for those offering analog functions or a mix of
analog and digital functions, they are made both as hybrid as w ~ l as l
MMICs.
6.21. Materials For Making MICs
The basic materials employed in fabrication of MICs, both hybrid
as well as MMICs, are classified into the following four categories :
1. Resistive Materials
2. Conductive Materials
3. Dielectric Materials
4. Substrate Materials

Resistive materials to be used for making MICs should desirably

have low temperature co-efficient of resistance, low voltage co-efficient
ofresistance, good stability at elevated temperatures and high humidity
conditions, low electrical noise and a range of values of sheet resis-
tivities typically in the range of 10 to 2000 ohms per square (alsowritten
as ohms /n ). The sheet resistivity is the ratio of volume resistivity to
sheet thickness. Commonly used resistive materials include Nickel-
Chromium, Tantalum Nitride, Tin oxide and Chromium silicide.
Conductive materials should desirably have high conductivity (i.e.
low sheet resistivity), good adherence to substrate material, good
solderability or bonding property, etchability, resistance to corrosion,
low temperature co-efficient of resistance and so on. Commonly used
conductive material for hybrid MICs is gold. For thin film metallisation
in MMICs, aluminium is used. Other materials in use are Silver,
Tantalum, Nickel and Chromium.
Dielectric materials mainlv used for real is in^ ca~acitorsand to
couple line structures should &sirably have low &el&ric loss, high
voltage withstanding capability, easy adaptability to various MIC
processes and stable dielectric constant. Commonly used dielectric
materials are Silicon monoxide and silicon dioxide used on silicon
wafers, Silicon nitride and Tantalum pentoxide.
Substrate matcrials should desirably have high purity, high
dielectric constant (determining the capacitive effects), low dielectric
loss, stable dielectric constant over the temperature and frequency
range of interest, high resistivity (determining the electrical insulation
between various elements), surface smoothness and high thermal con-
ductivity (determining the power dissipation capability). High purity
Alumina with polished surfaces, Quartz, Sapphire, Glass, Berrylia,
Gallium Arsenide, FerriteIGarnet are commonly used substrate
materials.
SEMICONDUCTOR MICROWAVE DEVICES 277

6.22. Thick Film Hybrid MICs

In Thick Film Hybrid MICs, passive components like resistors,
capacitors and interconnections are fabricated using thick film technol-
ogy and active devices like diodes, transistors and even monolithic chips
are added onto the substrate to get a fully functional hybrid MIC. The
thick film is obtained by what is known as thick film paste comprising
of a binder, usually glass frit, organic solvent, plasticisers and the
functional element which could be metal, alloy or ceramic depending
upon the functional requirement. The relative proportions of the binder
and the functional element determines the conductive, resistive,
dielectric properties of the film. The three basic components made by
using thick film are conductors (required for making interconnections,
as terminations for resistors, attachment pads for discrete devices to be
added, electrodes for thick film capacitorsj, resistors and dielectrics (as
crossover insulators for multilayer conductors, in thick film capacitors
etc.).
Processing
The two basic processes involved in thick film processing are :
1. Screening
2. Firing
The technique of screening in thick film hybrids is similar to silk
screening that we all know with the difference that silk screen is
replaced by a fine mesh like stainless steel screen and having mesh
counts in the range of 40 to 125 lineslcm. Line width as small as 100
microns is available. The screen is coated with a photoresist. The
desired pattern is formed on the photoresist by exposing it to ultraviolet
radiation selectively through a photo mask. The unexposed portions are
washed away with a developer solution. These are the portions cor-
responding to the pattern to be printed. The thick film is then forced
through the screen onto the substrate. The substrate is then air dried
by low temperature heat assisted drying. This is followed by firing
process which takes place in a furnace at a temperature in the range of
500 to 1000"C depending upon materials being used. The firing process
has three zones namely Preheat zone (250°C - 350°C),Hot zone (500°C
- 1000°C)and Cooling zone where the substrate is allowed to return to
room temperature with minimal thermal stress.
Thick Film Components
Resistors :Fig. 6.57 shows a typical thick film resistor geometry.
The sheet resistivity (or resistance) of this geometry will be (p/t) where
'p' is the volume resistivity or bulk resistivity. L I W is termed as the
Aspect Ratio. Resistance R of this thick film resistor is given by :
R = ps WWI
I t can be seen that resistance of a sheet like this will equal ps for
L = W irrespective of magnitude of L or W. That is, a lop x 10y square
will have the same resistance as a 100p x 1 0 0 ~ square. Thus the
278 MICROWAVES AND RADAR

resistance is given by sheet resistivity multiplied by number of squares

in the direction in which the resistance is being measured. Thick film
resistors are so designed that the aspect ratio is in the range of 25 : 1

to 1: 3 i.e. (1/3 < LfW < 25). A different sheet resistivity should be used
if necessitated by the required value of the resistance in order to keep
the aspect ratio within the preferredrange. Fig. 6.58 shows some typical
geometries for thick film resistors. These resistors need to be trimmed

Fig. 6.58
as they would have a tolerance as high as f 20% due to variations in
material, screened thickness, screenedgeometry,firing parameters and
so on. Trimming is done either by Abrasive trimming or by Laser
trimming. Some of the typical trim cuts are shown in Fig. 6.59.

Fig. 6.59
Capacitors :Thick film capacitors are fabricated using dielectric
pastes available in a wide range of dielectric constant. In case of
substrate capacitor (Fig. 6.60), the capacitor is formed across the
substrate with an electrode on one side and another electrode or a
ground plane on the other side with substrate acting as the dielectric.
Substrate capacitors are used for fabricating low value capacitors. Fig.
6.61 shows interdigitated type of capacitor usually available in the
range of 0.5 pF to 3 pF. A third type is parallel plate capacitor (Fig.
6.62). In a good design, the dielectric material should overlap the top
and bottom electrodes by 10 microns on all edges.
SEMICONDUCTOR MICROWAVE DEVICES 279

Substrate
Conductor
Ground Plane

Fig. 6.60

Electrode Electrode

Fig. 6.61

Conduc
Substrate
I
Fig. 6.62
6.23. Thin Film Hybrid MICs
In case of thin film hybrids, conductive, resistive and dielectric
films of relatively much lower thickness than thick film counterparts
are deposited or sputtered on an insulating substrate. The film thick-
ness typically varies from a few nanometers to a few microns. The film
thickness however becomes comparable to thick film thickness after
plating with the plating thickness in the range of 10 to 25 microns.
Different processes involved in the thin film processing are :

1. Vacuum Evaporation
2. Sputtering
3. Vapour Phase Deposition
4. Etching
5. Plating

The first three processes are for producing the thin film. One of
them is used for the purpose. While the advantage of vacuum evapora-
tion lies in its ability to deposit much thicker films in a reasonable time,
the films deposited by sputtering technique are much more uniform. In
addition, sputtering process allows different types of films of same
280 MICROWAVES AND RADAR

material having different characteristics. Vapour phase deposition is

similar to epitaxial process used for growing silicon crystal film and is
used for growing relatively thicker films.
Thc films are given desired patterns by the processes of
photolithography and etching. Plating is done to enhance conductivity
of films. It is done by either Electroplating or by Electroless plating.
Thin Film Components
Resistors :Thin film resistors have geometries similar to those
used in case of thick film resistors. Again sheet resistivity is the basic
design parameter. As mentioned abo~re,thin film is deposited over the
s u b s t r ~ t e~ n then
d given a conductive coating usins plating. Resistors
an6 condnctors are patterned using photolithography and etching tech-
niques. Tight tolerances can be obtained using laser trimming. Trim-
ming geometries are also similar to those explained in case of thick film
components. It is however not possible to have a wide range ofresistance
values. ?ower dissipation capability and voltage stress capability are
also -restricted, typical values being 10 W/sq. cm and 50 Vlmm.
Capacitors : Capacitors too can be of substrate type, inter-
di.gitated type or parallel plate type having geometries similar to the
on?s shown in case of tnick film capacitors.
6.24. Monolithic Microwave Integrated
Circuits (MMICs)
The Monolithic Microwave Integrated Circuits, theoretically, are
all those integrated circuits that work at microwave frequencies ir-
respective of whether they are analog or digital in nature. While analog
microwave circuits are usually designed for either sinusoidal signals or
a limited bandwidth in the microwave frequency range, the digital
circuits must be broadband as they have to handle relatively large
amplitude pulsed signals. In practice, it is generally the analog
microwave circuits that are designated as MMICs.
The technologies cormonly used far fabrication of MMICs in terms
of semiconductor materials and the basic active device used include :

1. Silicon bipolar fechnology using classical bipolar transistors

on silicon base
2 . GaAs MESFET technology using MESFET on GaAs base
3. AlGaAs / & A s HEMT technology
4 . SilGe bipolar technology where classical bipolar transistors
are replaced by heterojunction bipolar transistors (HBT) on
Si-Ge base.

Gallium Arsenide is the preferred substrate material and MES-

FET the chosen device technology for fabrication of MMICs. The salient
features of GaAs as a semiconductor material as a compared to silicon
SEMICONDUCTOR MICROWAVE DEVICES 281

which give it an edge over the latter for MMIC fabrication include the
following :
1.It offers a higher electron mobility. In fact, the electron velocity
passes through a value which is about three times the value for silicon
for electric fields in the region of 3 to 5 KVIcm before finally saturating
to a lower vaiue which is same as it is for silicon. Since the velocity in
the region under MESFET gate saturates only towards the end of the
region, the GaAs MESFET leads to devices with much higher speed.
2. Gallium Arsenide substrate is semi-insulating. Its resistivity is
much higher than that for silicon with the result that the interconnec-
tion lines made in GaAs have much lower losses.
3. GaAs is also much less sensitive to higher temperatures and
ionising radiation.
4. GaAs band gap corresponds to visible spectrum which allows
coupling of light emitting and light detecting diodes on the same chip.
5 . Its piezoelectric property allows coupling with surface acoustic
wave devices.
MESFET is ideally suited for many analog applications like
amplification, oscillation, phase shifting switching, frequency conver-
sion and so on. Schottky barrier diode is used as detector, mixer and
varactor. MESFET has anothcr advantage that it can also be used to
perform a diode's function to save processing steps as a MESFET and
a Schottky barrier diode need different doping concentration. This
reduction in complexity is, of course, a t the cost of slight reduction in
performance.

MMIC Fabrication Process

Fabrication of an MMIC involves a large number of processes
including creation of active layers, metallic layers and dielectric layers
having desired characteristics.
Active layers a n d Isolation :The process begins with a semi-in-
sulating GaAs substrate and formation of active layers. Active layer can
be grown by various methods. One of the methods is to grow the layer
by epitaxy using vapour phase deposition or molecular beam process.
It is necessary to grow an undoped buffer layer because impurities, if
any, in the substrate would lead to increased noise and other un-
desirable effects. It is particularly necessary when GaAs substrate is
doped with chromium to make the semi-insulating substrate. Active
layer with desired doping level is then grown on the buffer layer. Each
of the active layers meant for various devices to be integrated on the
chip needs to be insulated or isolated from the others. Isolation can be
accomplished by etching away the layers around the device or by boron
implantation. The advantage of implantation is that the surface
remains planar and also that isolationimplant does not need annealing.
Another method of cresting active layers is by ion implantation.
An advantage of ion implantation over epitaxial growth is that it can
282 MICROWAVES AND RADAR

be done locally using a mask of photoresist or a dielectric making

possible realisation of different doping concentrations on the same
substrate. But, ion implanted active layers have to be annealed at a
temperature of 700-800°C to repair the lattice damage caused by ion
bombardment.
Ohmic Contacts :The ohmic metal semiconductor contacts are
obtained by successive deposition of an alloy of 'Au-Ge', ' Ni ' and ' Au'
and then annealing a t 450°C.
Schottky Gates :Schottky gates are created by successive deposi-
tion of 'Ti', 'Pt' and 'Au'. 'Ti' gives the Schottky contact with the active
layer, 'Au' enhances conductivity and 'Pt' is used to prevent migration
of 'Au' into the active layer. Photolithography is used for deposition for
gate lengths larger than 0.5 micron. For gate lengths equal or smaller
than 0.5 micron, electron beam lithography is used.
Metallic and Dielectric Layers : First metallic layer that gives
lower electrodes of capacitors and some other connections is formed by
a deposit of 'Ti', 'Pt' and 'Au'. This is followed by formation of a dielectric
layer deposited to make dielectric of capacitors and protect active
devices. Silicon nitride is generally used for the purpose. A second
metallic layer of 'Au' is deposited to constitute upper electrode of
capacitors, spiral inductors and interconnections.
Air Bridges :Air bridges (Fig. 6.63) are used to make crossings
with low parasitic capacitances. These should be short to have good
mechanical strength. These are made by electroplating metal over a
photoresist stripe which is removed afterwards.

Fig. 6.63

Substrate

1 Ground Plane
Metallisation
I
Fig. 6.64.
SEMICONDUCTOR MICROWAVE DEVICES 283

Via Holes : Low inductance grounding, particularly of common

source FETs, is another in~portantrequirement of MMICs. I t is usually
done by etching a hole through the substrate: called a Via Hole, from
the back side and then metalising the hole and the back ground plane
together as shown in Fig. 6.64.
This necessitates making a photoresist pattern on the back side
that is aligned with the one on the front side. Alignment becomes easier
with a thinner substrate. Due to GaAs being a poor thermal conductor,
a thinner substrate offers an additional advantage of being a better heat
sink. However, a thinner substrate causes greater losses in microstrip
lines. The bath tub structure of Fig. 6.65 offers the advantages of both
thin and thick substrates. In this case, the substrate is thinned only a t
places where required.

Substrate

Fig. 6.65
Passive Components: Resistors are fabricated by using a length
of an active layer or a specific metallic deposit such a s that of 'Ti', 'Ni-Cr'
or 'TaN'. Capacitances are either of interdigitated type which gives low
value capacitors with good accuracy or of MIM-type (Metal-hsulator-
Metal). The dielectric used in MIM capacitors should have high permit-
tivity as well as high breakdown voltage. The figure-of-merit for the
dielectric is the dielectric efficiency expressed by :

Fig. 6.66
284 MICROWAVES AND RADAR

where 'E' is the breakdown electric field.

Inductors are of the spiral type as shown in Fig. 6.66. Air bridges
are used here to make interconnects. The transmission lines used in
MMICs to make interconnections and also for fabricating inductors are
the microstrip lines and other similar lines that adapt themselves well
to the MMIC structure.

Problem 6.17 : A thick film resistor is made from a material

having volume resistivity of 0.1 ohm-cm. Calculate its resistance if it has
a rectangular geometry with a n aspect ratio of 10 : 1 and thickness of
100 ym.
Solution : Resistivity, p = 0.1 R-cm
If (p,) is the sheet resistivity,
then p, = - P where t = thickness
t

If ( R )is the resistance value,

then R = ps x Aspect ratio
= 10x 10= loo n

Problem 6.18 : Refer to the two resistor geometries shown i n Figs.

6.67(a) and (b). The two resistors have been made from the same
material. If the resistance shown i n Fig. 6.67(a) has a value of 1000 Q
determine the sheet resistivity of the material used and also the resis-
tance value of the resistor whose geometry is shown i n Fig. 6.67 (b).
Solution :Resistance value (R)of a rectangular geometry resistor
is expressed as :
L
R=ps-
W
where ps = Sheet resistivity
L = Length
W = Width
For the resistor geometry of Fig. 6.67(a),

If (p) is the volume resistivity of the material used,

Then P = P s ~X t l
SEMICONDUCTOR MICROWAVE DEVICES 285

where t i = thickness of resistor geometry shown in Fig. 6.G7 ( a )

p=50x10x104=0.05~-cm

Therefore, ps2 =
P-- 0.05 = 25 n/a
tz 20 x
t2 = thickness of resistor geometry shown in Fig. 6.67 ( b )
ps, = Sheet resistivity of resistor geometry shown in Fig. 6.67 ( b )

(a) (b)
Fig.6.67
IfR2 is the resistance value of resistor shown in Fig. 6.67 ( b ) ,
then

Therefore. R? = 500 R.

Problem 6.19 :Calculate the capacitance of a thick film substrate

capacitor using a n alumina substrate having a relative permittivity of
9.6 and 100 ym x 100 pm square electrodes. Substrate thickness is 500
Pm.
Solution : Area of each electrode, A = 100 x 100 x 10-l2
-8 2
= 10 m

Substrate thickness, t = 500 x lo4

Relative permittivity, E, = 9.6
286 MICROWAVES AND RADAR
Problem6.20.Determine the resistance value ofa thin film resistor
having a geometry of FLE.6.68. The material used has a sheet resistivity

Fig. 6.68
Solution :Sheet resistivity, p, = 100 n/ r~
L 1.04 - 52
Number of squares = - = --
W 0.02
Therefore, resistance value, R = 100 x 52
= 5200 R
= 5.2 KQ.
SEMICONDUCTOR MICROWAVE DEVICES 287

Review Questions
1. Briefly describe the factors responsible for making bipolar
junction transistors unusable at microwave frequencies. What
advances have been made in the fabrication of these devices
to extend their usable frequency range ?
2. Explain the principle ofoperation of avaractor diode frequency
multiplier. How does i t differ from that of a step recovery diode
for the same application ? What would be the preferred con-
figuration if it is desired to multiply the source frequency of
about 150 MHz to more than 50 GHz ?
3. Why are parameteric amplifiers of low noise ? Differentiate
between degenerate and non-degenerate modes of operation
of parametric amplifiers ?
4. Briefly describejunction capacitance versus reverse bias volt-
age characteristics of a PIN Code. Define Dielectric relaxation
frequency and Cut-off frequency with reference to a PIN diode.
5. Briefly describe the use of a PIN Diode as (i) Attenuator (ii)
Switch (iii) Phase shifter
6. Draw the V-I characteristics of a tunnel diode and explain
different regions. What is a Backward diode ?
7. What is Transferred Electron Effect ? What are the semicon-
ductor material characteristics required for it to exhibit this
effect ?
8. Briefly describe the Gunn mode and Limited Space Charge
Accumulation Mode in case of Gunn diodes.
9. What are transit time devices ? Briefly describe the principle
of operation of an IMPATT diode.
10. Differentiate between MESFET a n d HEMT types of
microwave FETs. Compre their advantges and disadvantages.
11. Briefly describe the principle of operatioin of a Ruby maser.
What is responsible for making maser a low noise amplifier ?
12. What does acronym LASER stand for ? How does a laser
function ? What are the chief attributes of a laser device ?

Problems
1. What can be the minimum allowable load conductance if the
magnitude of negative differential conductance of the device
used to amplify a microwave signal from a source and feed it
to the load is 0.01 mho.
10.0025 mhol
2. A Barritt diode oscillator employs 5 micron wide N-region.
Determine the oscillator output frequency if the saturated
velocity of charge carriers is lo5 m/s.
[15 GHzl
3. A Certain GaAs semiconductor material is characterised by :
288 MICROWAVES AND RADAR

Relative permittivity = 15
Mobility = 8500 cm2/V-s
Doping concentration = 1015/cm 3
Compute its dielectric relaxation time.
[0.97 ps]
4. A microwave bipolar tansistor is observed to have an ( ~ F Eof)
8 a t an operational fiequency of 5 GHz. What should be its
(f?) specificaSon ?
[40 GHzl
5. A laser is found to emit a t 1micron. If the energy of the lower
laser level is 4 x Joule. I?etermine the upper laser level
energy in (eV) given that p:ancks constant = 6.63 x
Joule-second.
[ 1.5 eVl
6. A Barritt diode has a 15 micron wide active layer. If the
maximum electric field along tne length of ative layer is 150
KV/cm, detrmine its punch through voltage., V ~ T
1225 V

SELF EVALUATION EXERCISE

Multichoice Questions
One of the following is an active microwave device
(a)Strip line (b)Microstrip (c)SAW device (d)IMPATT
One of the following microwave devices is based on the trans-
ferred electron effect
(a)Gunn diode ( b )BARlTT diode
(c) IMPATT diode ( d )Tunnel diode
One of the following semiconductor materials does not exhibit
Gunn effect
(a)Si ( b )G ~ I A ~ (c) InP (d)GaAsP
The commonly used semiconductor material for fabrication of
MESFET is
(a)GaAs (b)Si (c) Ge (d)InP
One of the following microwave devices is used for switching
applications.
(a)Gunn diode ( b )PIN diode
SEMICONDUCTOR MICROWAVE DEVICES 289

( c ) Step recovery diode ( d )IMPATT diode

6 . One of the following devices is better suited to fabrication of
MMICs.
( a )GaAs MESFET ( b )IMPATT
( c ) Bipolar transistor ( d )Gunn diode
7. With reference to thin film and thick film technologies, sheet
resistivity
( a )is ratio of volume or bulk resistivity to film thickness
( b )is same as volume or bulk resistivity
( c )has the units of ohms per cm
(dl when divided by the aspect ratio gives the resistance value
8. One of the following is not an advantage for preferring GaAs
over silicon for fabricating monolithic microwave integrated
circuits.
( a )GaAs offers highest electron velocity
(b) GaAs technology is very well established
( c ) GaAs has much higher resistivity
( d )GaAs is much less sensitive to ionising radiation
9. One of the following is not the reason for silicon being totally
unsuited for fabrication of Gunn diode.
( a ) The electron velocity in silicon varies very slowly as a
function of applied electric field
( b ) Silicon has a relatively lower resistivity as compared to
Gallium Arsenide
( c ) Silicon does not exhibit negative differential mobility as
a function of applied electric field
(d) Silicon is sensitive to ionising radiation
10. Punch Through Voltage is associated with
( a )IMPATT diode (b) TRAPATT diode
(c) Gunn diode (d)PIN diode.

11. Below are given brief descriptions of some of the popular

microwave devices and circuits. You have to identify the
relevant device or circuit for each of these descriptions.

1. This microwave device is commonly used as a low level, low

noise microwave oscillator and it depends for its operation on
the negative differential carrier mobility when the applied
electric field exceeds a certain threshold.
290 MICROWAVES AND RADAR

2. It is a microwave,diode in which rarric :r generation is due to

an avlanche mwhanism and also :he ti ansit time through the
drift space is vital for generation of oscillations.
3. It is the preferred device for making MMICs.
4. It is used for extremely low noise amplification of microwave
signals. It needs another source of microwave signal also for
operation.
5. It is used in microwave frequency multipliers and it is not a
varactor diode.
6 . It can be used for generation of microwave oscillations when
placed in a resonant cavity and has a similar device operating
a t optical frequencies called the Laser.

111. Fig. 6.69 (a)to (e) show performance characteristics

related to some of the common and well known semiconductor
microwave devices. Identify the devices.

(c) (4
Fig. 6.69(Contd.)
SEMICONDUCTOR MICROWAVE DEVICES

(d
Fig. 6.69
 Antennas
n antenna is basically a structure that acts as an interface to

A couple the output of the transmitter or the input of the receiver,

as the case may be, to free space. It i- a vital link in the chain
of systems and subsystems constituting an entire communica-
tion set-up. It does the job of transmitting high frequency electrical
currents fed to it from the output of the transmitter into electromagnetic
waves in case of transmitting antenna and that of intercepting
electromagnetic waves and generating an equivalent electrical signal
in case of a receiving antenna. It can also be thought of as a transducer
that converts guided electromagneticwaves into free space electromag-
netic waves and vice-versa. The present chapter begins with an over-
view of antenna fundamentals and the basic concepts relevant to their
functioning. Concepts like radiation mechanism, reciprocity etc. have
been briefly discussed. This is followed by definition and interpretation
of antenna parameters including Directive gain, Power gain, Input
impedance, Bandwidth, Beamwidth, Effective aperture, Directional
pattern, Side lobe level, Polarisation and so on. Different types of
antennas are discussed next with emphasis on microwave antennas. A
large number of illustrative examples have also been included. A self
evaluation exercise is given towards the end of the chapter.
The topics covered in this chapter include :
What is an ante- 7.9. Yagi Antenna
nna 7.10. V a n d Rhombic
Antenna Reci- Antenna
procity 7.11. Reflector Anten-
Radiation Mecha- nas
nism 7.12. Lens Antennas
Antenna Para- 7.13. Horn Antenna
meters
Resonant and 7.14. Helical Antenna
Non-resonant An- 7.15. Log-Periodic An-
tennas tenna
Electrical and 7.16. Phased Array An-
Physical Lengths tenna
Types of Antennas 7.17. Microstrip An-
Hertz., Di ole and tenna
In
Marconi tenna
ANTENNAS 293

7.1. What is an Antenna !

An antenna, as mentioned above in the introductory paragraph, is
a structure that transforms guided electromagnetic waves into free
space electromagnetic waves and vice-versa. It is defined as the struc-
ture associated with the region of transition between a guided
electromagnetic wave and a free space wave or vice-versa. The guided
electromagnetic waves would look more appropriate when the feeder
connecting the output of the transmitter and the antenna or the input
of the receiver and the antenna is a waveguide which is generally true
when we talk about microwave frequencies and microwave antennas.
In case of other antennas such as those at HF, VHF, the term guided
electromagneticwaves mentioned above would be interpreted as guided
electromagnetic signal in the form of current and voltage. According to
some books, antenna is considered as a system that comprises of
everything that is connected between the transmitter output or the
receiver input and free space. This includes, in addition to the com-
ponent that radiates, other components such as feeder line, balancing
transformers etc.

7.2. Antenna Reciprocity

An antenna is a reciprocal device, that is, its directional pattern
as receiving antenna is identical to its directional pattern when the
same is used as a transmitting antenna provided, of course, it does not
employ unilateral and non-linear devices such as some ferrites. Also,
reciprociQ applies provided that the transmission medium is isotropic
and the antennas remain in place with only their transmit and receive
functions interchanged. Antenna reciprocity also does not imply that
antenna current distribution is the same on transmission as it is on
reception.

7.3. Radiation Mechanism

When an RF signal is applied t o the antenna input, there is a
current and voltage distribution on the antenna which lead to existence
of an electric and a magnetic field. The electric field reaches its maxi*
mum coincident with the peak value of the voltage waveform as shown
in Fig. 7.l(a).If the frequency of the applied RF input is very high, the
electric field does not collapse to zero as the voltage goes to zero. Alarge
electric field is still present. During the next cycle, when the electric
field builds up again, the previously sustained electricfield gets repelled
from the newly developed field. This phenomenon is repeated again and
again and we get series of detached electricfields moving outwards .from
the antenna.
According to laws of electromagneticinduction, a changing electric
field produces a magnetic field and a changing magnetic field produces
an electric field. It can be noticed that when the electric field is at its
maximum, its rate of change is zero and when the electric field is zero,
its rate of change is maximum. This implies that the magnetic field's
maximum and zero points correspond to the electric field's zero and
294 MICROWAVES AND RADAR

Magnetic field

(a)
H

Direction of ' d

H
(b)
@ Out o f plane of paper @ Into plane of paper
- 7.1
Fig.
maximum points. That is, the electric and magnetic fields are at right
angles to each other and ~o are the detached electric and magnetic fields.
The two fields add vectorially to give one field which travels in a
direction perpenriicular to the plane carrying mutually perpendicular
electric and magnetic field as shown in Fig. 7 . l ( b ) .
7.4. Antenna Parameters and Characteristics
The undermentioned parameters are briefly described in the fol-
lowing paragraphs :
1. Directive Gain
2. Power Gain
3. Directional Pattern
4. Beam Width
5. Bandwidth
6. Polarisation
7. Impedance
7. Aperture
Directive Gain
An antenna is considered to have a gain in the sense that it
concentrates the radiated energy in a certain desired directionls instead
of radiating uniformly in all directions with the result that the power
ANTENNAS 295

Directive Gain
An antenna is considered to have a gain in the sense that it
concentrates the radiated energy in a certain desired directionh instead
of radiating uniformly in all directions with the result that the power
density in that specific direction has to be greater than it would be, had
the antenna been an isotropic radiator. An isotropic radiator is a
hypothetical antenna radiating equally in all directions i.e. having a
spherical radiation pattern. It is considered as a standard reference
antenna. So, when we talk about the gain of an antenna, we basically
talk about its directive gain.
The Directive Gain in a given direction is defined as the ratio of
the power density of the radiated electromagnetic energy in that direc-
tion to the power density in the same direction and at the same distance
due to an isotropicradiator with both antennas radiating the same total
power. Fig. 7.2(a)shows the directional pattern of an isotropic radiator.
The directional patterns of two other antennas are shown in Figs. 7.2
( b ) and 7.2(c). Clearly, the antenna represented by the directional
pattern of Fig. 7.2(c)has a higher directive gain.

w
(a) (b) (c)
Fig. 7.2.
The directive gain is therefore nothing but ability of the antenna
to concentrate the radiated energy in a given direction. As is clear from
the above description, directive gain is always specified for a given
direction and would have maximum value in the direction of maximum
radiation. This maximum directive gain is termed as theDirectivity and
is usually expressed in decibels (dB). An antenna having a Directivity
of 20 dB would produce a power density at a given distance in the
direction of maximum radiation when radiating a certain totd power
that would be 100 times the power density resulting from an isotropic
radiator at the same point when radiating the same total power.
Another similar term related to the directivity of the antenna is the
power gain described below.
Power Gain
Definition of Power Gain is similar to that of Directive Gain or
Directivity except that it is not the power radiated by the antenna but
the power fed to the antenna that is considered while computing the
gain. It takes into account the antenna losses and thus is of greater
practical importance. It may be defined as the ratio of the power density
at a given distance in the direction of maximum radiation intensity to
296 MICROWAVES AND RADAR

the power density at the same distance due to an isotropic radiator for
I
the same total power fed to the two antennas.
Directional Pattern
The Antenna Directional Patter12 or Radiation Pattern is a nor-
malised plot of distribution of electromagnetic energy in three dimen-
sional angular space. The parameters to be plotted could be radiation
intensity which is the power per unit solid angle or the power density.

Fig. 7.3
Fig. 7.3. shows the typical three dimensional directional pattern
ofan antenna. The power density in dB is plotted vertically in the plane
of paper versus the azimuth and elevation angles plotted alongmutual-
ly perpendicular axes in rectangular coordinates. More commonly used
representation of directional pattern is the two dimensional plot. There
are again various types of 2-D plots. One of the types is the polar plot

Fig. 7.4
ANTENNAS 297

of radiation intensity or power density that we are all quite familiar

with (Fig. 7.4). Another is principal plane elevation pattern as shown
in Fig. 7.5. This is the pattern drawn by sectioning the 3-D pattern with
a vertical plane through the peak of the beam and zero azimuth angle.
A similar pattern called the principal plane azimuth pattern couldvbe
drawn by sectioning the 3-D pattern through the peak of the beam and
zero elevation angle. Though the 2-D patterns obtained by sectioning
with planes 4 e r than the principal planes (called cardinal planes) can
also be drawn, the azimuth and elevation patterns usually suffice in
most of the cases.

Elevation (Deg) -
Fig. 7.5
The main beam of the pattern is called the mainlobe and the beams
in directions other than the direction of maximum radiation are called
side lobes. High side lobe level, with a few exceptions, are always
undesirable. The side lobe level of an antenna pattern is usually
specified in terms of relative side lobe level which is the peak level of
the highest side lobe relative to the peak of the main lobe. For instance,
a relative side lobe level of - 20 dB means that the peak power density
in the side lobe is l / l O o t h of the peak power density in the main lobe.
The common types of antenna radiation patterns include ( a )Omni-
directional (azimuthal plane) beam ( b ) Pencil beam ( c ) Fan beam and
( d )Shaped beam.
The omni-directional beam is commonly used in communication
and broadcast applications for obvious reasons. The azimuth plane
pattern is circular and the elevation pattern has some directivity to
increase the gain in horizontal directions.
Pencil beam is highly directive pattern whose main lobe is confined
to within a cone of small solid angle and is circularly symmetric about
the direction of maximum intensity.
A Fan beam is narrow in one direction and wide in the other. A
typical application of such a pattern would be in search or surveillance
298 MICROWAVES AND RADAR

radars in which wider dimension would be vertical and the beam is

scanned in azimuth. The other application would be in height finding
radar where the wider dimension is in the horizontal plane and the
beam is scanned in elevation.
There are applications that impose beam shapingrequirements on
the antenna. One such requirement, for instance, is to have a narrow
beam in azimuth and a shaped beam in the elevation such as in case of
air search radar.
Beam Width
Beam width gives the angular characteristics of its radiation
pattern. It is taken as the angular separation either between the half
power points on its power densiiy radiation pattern (Fig. 7 . 6 ~ or )
between - 3 dB points on the field intensity radiation pattern (Fig. 7.6b).
It is measured in degrees and with reference to the main lobe. Antennas
also have 6 dB beam widths and null-to-null beam widths. Null-to-Null
beam width is width of the response between the minima surrounding
the main lobe and is approximately twice the 3 dB beam width for most
of the antenna responses. The parameter is particularly relevant to the
antennas producing narrow beams such as those in tracking radars.
Antenna's power gain G (8,4) is related to its beam width parameters
by :

(b)
Fig. 7.6
ANTENNAS

where R = Solid angle (in steradians) = MI . A@

with A0 = Beam width in azimuth direction (in radians)
A$ = Beam width in elevation direction (in radians)
Bandwidth
Antenna Bandwidth is in general the operating frequency range
over which the antenna gives a certain specified performance. Antenna
bandwidth is always defined with reference to a certain parameter such
as gain or input impedance or SWR. It is generally taken as the
frequency range around the nominal center frequency over which power
gain falls to half of the maximum value. When referenced to SWR, one
may specify a 2:l SWR bandwidth and so on.
Lower the operating frequency, narrower is the bandwidth. It
follows from the rule that in case of a resonant circuit, for a given 'Q',
the bandwidth is directly proportional to the center frequency.
Polarisation
Antenna polarisation i s the direction of electric field vector with
reference to ground in the radiated electromagnetic wave while trans-
mitting and the orientation of the electromagnetic wave again in terms
of direction of electric field vector the antenna responds best to while
receiving. From antenna reciprocity, we can say that the antenna would
respond most optimally to an electromagnetic wave that would have
the same polarisation as that of the transmitted wave radiated from the
same antenna. It is a normal practice to consider the antenna itself as
being polarised. The polarisation of the antenna is the same as that of
the electromagnetic wave it radiates or best responds to.
The polarisation of an antenna can be classified into two broad
categories of Linear Polarisation and Elliptical Polarisation. Linear
polarisation could be either Horizontal Polarisation or Vertical
Polarisation. Circular Polarisation is a special case of elliptical polarisa-
tion.
In Linear Polarisation, the electric vector lies in a plane, If the
plane is horizontal, i t is horizontally polarised and if the plane is
vertical, it is vertically polarised. Inclined plane leads to what may be
referred to as Slant Polarisation. Slant polarisation is a general case of
linear polarisation having both horizontal and vertical components. It
is called linear polarisation because the direction of the resultant
E-vector is constant with respect to time. In the generalised case of a
linearly polarised wave, the two mutually perpendicular components of
the E-vector are in phase.
When the two components of the E-vector are not in phase. it can
be verified that the tip of the resultant traverses an ellipse as the RF
signal goes through one complete cycle. This is called elliptical polarisa-
tion. This polarisation could have right hand sense or left hand sense
depending upon whether the E-vector moves clockwise or anticlockwise
when viewed as a wave receding from the observation point in the
300 MICROWAVES AND RADAR

direction of propagation. Elliptical polarisation has two orthogonal

linearly polarised components. When the magnitudes of these com-
ponents become equal and the phase difference between the two be-
comes 90°, the polarisation becomes Circular Polarisation. Again, we
have either Right Hand Circular Polarisation (RHCP) or Left Hand
Circular Polarisation (LHCP). The three types of polarisation namely
Linear, Elliptical and Circular are shown in Fig. 7.7.

Fig. 7.7
Cross-Polarisation:Cross-polarisation is that component which
is orthogonal to the desired polarisation. For instance, a horizontally
polarised antenna may also radiate vertical polarisation in some direc-
tions of propagation or a vertically polarised antenna may radiate
horizontal polarisation in some directions. The other example could be
that of RHCP antenna also radiating LHCP and an LHCP antenna also
radiating RHCP. A well designed antenna should have a cross-polarised
component at least 20 dB below the desired polarisation in the direction
of the main lobe and 5 to 10 dB below the desired polarisation in the
direction of sidelobes.
Polarisation Loss :If the received electromagnetic wave is of a
polarisation different from the one the antenna is designed for, a
polarisation loss results. This loss in dB in case of linear polarisation is
given by :
Loss = 20 Log [ l/cos $1
where 4 = angle between the polarisation of the received wave and that
of the antenna
ANTENNAS 301

Table 7.1

Loss for the cross polarised component would be infinity as ($J)

equals 90'. Also, loss for the received cikular polarisation for a linearly
polarised antenna would be 3 dB as half of power resides in each of the
orthogonal linear components constituting circular polarisation. Also
in circular polarisation, the E-vector would always make an angle of
45" with either of the two linear components. The polarisation loss for
received LHCP wave for an RHCP antenna would be infinity and
vice-versa as the two can mathematically be proved to be orthogonal.
Choice of polarisation is a very important design parameter. For
instance, circular polarisation is used to suppress rain clutter in radars
because an RHCP wave after reflection from rain droplets becomes
LHCP and vice-versa and any reflections from rain will not thus be
received. Similarly, horizontally polarised jamming signal can be best
countered by employing a vertically polarised antenna. In fact, many
radar and communication systems have polarisation diversity. Txble
7.1 below shows polarisation loss for various combinations of antenna
and received electromagnetic wave polarisations.
Antenna Impedance
The Antenna Impedance at a given point in the antenna is given
by the ratio of voltage to current at that point. As the magnitude of
voltage and current vary along the antenna length, the impedance also
varies being minimum at the point of voltage node or minima such as
the center point ofhalf wave dipole and maximum at the point of current
node such as the center point of full wave length long antenna. The
input impedance of an antenna is of considerable importance to en-
gineers as it is desirable to supply the maximum amount of transmitter
power to the antenna. FOPthis, the characteristic impedance of the
feeder line must match the antenna input impedance at the chosen feed
point.
The antenna impedance is resistive if it is resonant at the operat-
ing frequency. The antenna resistance further comprises of two com-
ponents namely the'Radiation Resistance and the Loss Resistance.
Radiation resistance is basically that resistance which if the antenna
302 MICROWAVES AND RADAR

is terminated with would dissipate the same power as that being

radiated by the antenna. It is given by the radiated power divided by
square of feed current. The loss resistance is contributed by factors like
eddy current losses in metallic objects lying in the vicinity of induction
field of antenna, losses in imperfect d;elsctrics,corona effect etc. Anten-
na efficiency is defined as :
q = [Rr/!Rr + Rd)l
For short dipoles whose effective length is less than say (h/8), the
radiation resistance can be computed to a good approximation from :

-
Antenna Aperture Effective Aperture -Aperture
Effectiveness
TheAntennaApertureis the physical area of the antenna projected
on a plane perpendicular to the direction of the main beam or the main
lobe. In case of main beam axis being parallel to the principal axis of
the antenna, it is the same as the physical aperture of the antenna itself.
For a given antenna apertureA, the directive gain of the antenna
at an operating wavelength of 'h' is &ven by :

This expression is valid only when the aperture A is uniformly

illuminated. Typical antennas are not uniformly illuminated and have
a tapered illumination being m&mum at the center and lower towards
the edges. This is done to redwe the side lobe level. Due to this
non-uniform illumination, the antenna gain falls from its maximum
value of [4d/h]. This is where the term Effective Aperture of the
antenna comes into picture. It is that aperture area which when
uniformly illuminated gives the same gain as that offered by a non-
uniformly illuminated a n t e n ~ of
a aperture A. Thus gain of a practical
antenna is given by :

where (A) and (Ae) are interrelated by :

A=q.Ae
q = Aperture Efficiency (or Effectiveness)
(q) is again product of length efficiency in azimuth and length
efficiency in elevation directions and is typically 0.5.

Problem 7.1 :Calculate the 'Q'of a 50 cm long half wave dipole

having a bandwidth of 10 MHz.
Solution :The length of the dipole = h/2 = 50 cm
Therefore, h = 100 cm
The operating frequency can now be computed from
ANTENNAS 303

f, = c/h = 3 x 1 0 ~ ~ / 1 0
= 0300MHz
The 'Q' of the antenna can be computed from :
Q = Centre Frequency/Bandwidth = 300/10 = 30

Problem 7.2. An antenna has a radiation resistance and loss

resistance of 72 and 8 ohms respectively. If the antenna has apowergain
of 27, determine its Directivity in dB.
Solution :If (Ap)and (D) and (q) are respectively the power gain,
the directivity and the radiation efficiency of the given antenna, then
Ap = q.D i.e. D =Ap/q
Also, q = Rr/(Rr + Rd) where Rr = Radiation resistance and Rd =
Loss resistance
Therefore, q = 72/(72 + 8) = 0.9 which gives:
D = 27/0.9 = 30, or D (in d B ) = 10 log 30 = 14.77 dB

Problem 7.3 : A tracking radar antenna produces a pencil like

beam with both azimuth and elevation beam width equal to 0.5' each.
Determine the gain of the antenna in dB. Also calculate the antenna
aperture if the radar wavelength is 3 cm.
Solution :Azimuth beam width, A8 = 0.5' = 0.5W180 radian =
0.00872radian
Elevation beam width, A+ = 0.5" = 0.00872 radian
Antenna gain, G = 4dA8 . A+ = 4~/(0.00872x 0.00872)
= 165180 = 52.18 dB (= 10 log 165180)
Radar emission wavelength = 3 cm
Also, antenna gain can be expressed in terms of antenna aperture,
(A) by :
G = 4 d / h 2 which gives A = ~ h ~ / 4 =
r c165180 x 0.03 x 0.03/4n
= 11.836 m

Problem 7.4: A cosecant squared receivingantenna has an actual

projected area to the received beam of 10 square meters. The main lobe
of its directional pattern has a length eficiency of 0.7in the elevation
and 0.5 in the azimuth directions. Determine the effectiveaperture of the
antenna.
Solution : Length efficiency in azimuth direction = 0.5
Length efficiency in elevation direction = 0.7
Therefore Aperture efficiency, q = 0.5 x 0.7 = 0.35
Effective Aperture,Ae = q. A = 0.35 x 10 = 3.5 sq.m.
304 MICROWAVES AND RADAR

Problem 7.5 : A certain antenna radiates a total power of 100 W

to produce a power density of 10 mW/sq.m at a distance of 1 k m from
the antenna in the direction ofits maximum radiation. It is also observed
that the same power density can be produced at the same point when the
antenna is replaced by a n isotropic radiator radiating a total power of
10 KW. Determine the Directivity of the antenna i n decibels.
Solution : Directivity (in dB) = 10 log [ 10000 / 100 1 = 10 log
100 = 20 dB

Problem 7.6 : A 3 m diameter circular reflector antenna operates

at 10 GHz. Its length efficiency is 0.7 and its radiation efficiency is 0.9.
Determine its Beam width, Effective aperture and Gain.
Solution :Diameter of antenna = 3 m
Length efficiency = 0.7
Therefore, Effective diameter, Deff = 3 x 0.7 = 2.1 sq.m
Beam width = h/Deff where h = 3 x 1010/1010= 3 cm = 0.03 m
Beam width = 0.03/2.1= 0.0142 radian = 0.81"
The azimuth and elevation length efficiency in case of circular
reflector antennas is same.
Therefore, Aperture efficiency = 0.7 x 0.7 = 0.49
Actual area = n x 3 x 3/4 = 7 sq.m
Therefore, Effective aperture = 7 x 0.49 = 3.43 sq.m
Gain = [ 4 n ~ e / h ' ] = [4 x 3.14 x 3.43/0.03~]= 47870 = 46.8 dB

Problem 7.7: Compute the radiation resistance of a @/lo) wire

dipole i n Fee space.
Solution. Radiation resistance.of short dipoles is given by
790 [l/h12
Rr = 790 [ ?dl0 h12 = 7.9 ohms

Problem 7.8 : A certain rectangular antenna is 6 m wide and 4

m high. It has a n azimuth length efficiency of 0.7 and elevation length
efficiency of 0.5. Compute its azimuth and elevation beam widths,
effective aperture and gain in d B if the operating wavelength is 3 cm.
Solution : Azimuth length = 6 m
Effective azimuth length = 6 x 0.7 = 4.2 m
Elevation length = 4 m
Effective elevation length = 4 x 0.5 = 2 m
Actual area = 6 x 4 = 24 sq.m
Aperture efficiency = 0.7 x 0.5 = 0.35, which gives
ANTENNAS 305

Effective aperture = 24 x 0.35 = 7.4 sq.m

Azimuth beam width = [WEffectivelength in azimuth]
=[0.03/4.2] = 0.00714 radian = 0.41"
Elevation beam width = [WEffectivelength in elevation]
= [0.03/2] = 0.015 radian = 0.86'

Gain = [41c/h2].~e= [4 x 3.14/0.03 x 0.031 x 7.4 = 117226

Problem 7.9 : Determine the beam width between nulls of a

paraboloid reflector antenna having a 3 d B beam width o f 0.4" and an
effective aperture of 5 sq.m.
Solution : The null-to-null beam width of most of the antenna
responses is twice the 3 dB beam width.
Therefore, null-to-null beam width = 2 x 0.4 = 0.8"

Problem 7.10. The received signal strength i n a certain horizon-

tally polarised antenna is 20 d B when receiving Right Hand Circularly
Polarised (RHCP) electromagnetic wave. Computer the received signal
strength when
( a ) the incident wave is horizontally polarised
( b ) the incident wave is vertically polarised
( c ) the incident wave is left hand circularly polarised (LHCP)
( d ) the received wave polarisation is making a n angle of 60' with
the horizontal
Solution : When the incident polarisation is circularly polarised
and the antenna is linearly polarised, there is a polarisation loss of 3
dB.
Therefore, incident signal strength = 20 + 3 = 23 dB
( a ) When the received polarisation is same as the antenna
polarisation, the polarisation loss is zero.
Therefore, received signal strength = 23 dB
( b ) When the incident wave is vertically polarised, the angle
between the incident polarisation and the antenna polarisa-
tion is 90°, therefore
Polarisation Loss = 30 log [l/cos 41
= 20 log [1/0]

Therefore, received signal strength = zero

( c ) When the incident wave is LHCP and the antenna polarisation
is linear, there will be a 3 dB polarisation loss and the received
signal strength therefore will be 20 dB only.
( d ) In this case, the angle between the incident wave polarisation
and the antenna polarisation is 60".
306 MICROWAVES AND RADAR

Polarisation Loss = 20 log [l/cos I$1 = 20 log [l/cos 60°]

= 20 log 2 = 20 x 0.3010
=6dB
Therefore, received signal strength = 23 - 6 = 17dB
7.5. Resonant and Non-resonant Antennas
A resonant antenna is characterised by the following features. Its
length is an integral multiple of half the wavelength and there are both
forward as well as reflected waves on the antenna leading to existence
of standing waves. A resonant antenna is analogous in behaviour to a
resonant transmission line. The directional pattern of a resonant an-
tenna is always bi-directional due to presence of both forward and
reflected waves. Fig. 7.8 shows directional patterns of h/2, h and
3h/2 long antennas. In general, number of lobes on one side in the
directional pattern of a resonant antenna equals the number of half
wavelengths in the antenna length.
A non-resonant antenna on the other hand has no standing waves.
It has only the forward travelling waves and no reflected waves. This
is achieved by terminating the antenna with a suitable load resistance
which ensures that there is no reflection of energy. A rhombic antenna
is a typical example of a non-resonant antenna. Non-resonant antennas
have a unidirectional pattern. Fig. 7.9 shows the rhombus antenna and
its typical directional pattern.

Dipole ' , D~pole \ / Dipole

..d' , ,.-' . ..-,
I

Fig. 7.8

Fig.7.9
ANTENNAS 307

7.6. Electrical and Physical Length

An antenna is observed to behave as if it was larger than its
physical length. This effective length, also called the electrical length,
is larger than the physical length due to presence of some parasitic
effects such as the finite thickness of antenna conductor and end effects.
The amount oflengthening dependsin a complicated manner on various
parameters including the characteristic impedance, the length and the
configuration of the antenna. This effect is somewhat greater for anten-
nas having a lower characteristic impedance than for thin wire anten-
nas. As a rule of thumb, the physical length of the antenna is made
about 5% shorter than the desired length to allow for this increase so
that the effective length becomes the desired length.

Problem 7.11 : Calculate the length of a halfbave dipole to be cut

so as to be resonant at 300MH.z after allowing for required compensation
due to parasitic effects.
Solution : Operating frequency = 300 MHz
Wavelength, h = c / f = 3 x 10'/300 x lo6 = 1m
h / 2 = 1/2 m = 50 cm
Therefore, length to be cut = 50 - 0.05 x 50 = 47.5 cm

7.7. Types of Antennas

In the paragraphs t o follow, we shall describe the operational
aspects of major antenna types. The types other than those for
microwave applications will only be briefly described. Microwave an-
tennas will be discussed at length touching upon both their functional
features as well as the design aspects. The types to be discussedinclude
the following :
( a ) Hertz and Marconi antennas
( b ) Yagi-Uda antenna
( c ) Rhombus antenna
( dj Reflector antenna
( e ) Lens antenna
(f) Hornantenna
(g) Helical antenna
(h) Lon periodic antenna

(j) Microstrip antenna

7.8. Hertz, Dipole and Marconi Antennas
A Hertz Antenna is a straight length of a conductor that is half
wave long. It may be placed vertically to produce vertically polarised
waves (Fig. 7.10~)or in horizontal position to produce horizontally
polarised waves (Fig. 7.10b).
308 MICROWAVES AND RADAR

A Dipole Antenna. is also a straigh rad:ator usually fed at t5e

center and producing -maximum of rauiaiion in a place prrpendicular
to the antenna axis. A dipole that is half wave long is called Half jVave
Dipole (Fig. 7.10~).
The vertical antenna that is quarter w.lve long and is fed against
an infinitely large perfectly conducting plane is called the Quarter Wave
Monopole or MarconiAntenna. It has the same radiation characteristics
above the plane as does the half wave dipole a n t e ~ n in
a free space. This:

(a)
(d)
Fig. 7.10
is due to the reason that the fields when reflected from the plane appear
to originate at an image element located beneath the plane. The
Marconi antenna has an edge over the Hertz antenna when it is to be
used as a transmitting antenna at low frequencies as its length is half
of the required length of Kertz antenna for a given transmission
frequency. Also, a Marconi antenna produces vertically polarised
waves, ideally suited for transmission and propagation of relatively
lower frequency RF signals. The radiation resistance ofhalf wave dipole
and the quarter wave monopole can be determined to be equal to 73
ohms and 36.5 ohms respectively. The generalised expre~sionsfor the
impedance are :
0.609 q hs2
Impedance (Monopole)=
2n
0.60911 I?,
Impedance (Half Wave dipole) =
n
where Irm = RMS value of antenna current
q = Characteristic impedance of the medium
= 377 ohms for free space
A modification of the half wave dipole is the folded dipole suitable
for TV reception purpose. A folded dipole (Fig. 7.10d) comprises of two
half wave dipoles connected at the ends and one of them fed at the
center. It may be constructed by folding a full wave long conductor. The
second element gets its excitation from the field produced by the driven
element. The folded dipole electrically behaves in the same fashion as
ANTENNAS 309

a straight dipole, physical construction being the only difference. Addi-

tion of this second element increases the input impedance of the
antenna which is given by :
cross-sectional area of all conductors
ztf=ziX cross-sectional area of driven element I2

where Z i f = Input impedance of folded dipole

Zi = Input impedance of straight dipole
Thus, if the second element is of the same cross-section as the
driven element, the input impedance increases by a factor of 4. A folded
dipole thus has an input impedance of approximately 300 ohms and at
the same time having same directional characteristics as the conven-
tional half wave dipole. This impedance matches the characteristic
impedance of the twin lead feeder, the commonly used transmission line
for TV reception purpose.
8.8. Yagi-Uda Antenna
A Yagi-Uda antenna comprises of a half wave dipole with parasitic
elements to enhance the directionality of the radiation pattern. It is the
most commonly used antenna type for HF and VHF communications.
The simplest Yagi antenna would be a 3-element array having a center
fed half wave dipole as the driven element, one parasitic element
smaller in length than the driven element by about 4% and called the
Director placed in front of the driven element and another parasitic
element larger in length than the driven element by about 5%, called
the Reflector, and placed behind the driven element (Fig.7.11). The
director enhances directivity of the radiation pattern and the reflector
suppresses radiation in the backward direction i.e. when used as a
receiving antenna, it does not receive from that direction thus impro-
ving the front-to-back ratio. The director-dipole spacing is approximate-
ly 0.12 h where as the reflector-dipole spacing is 0.2h. Fig. 7.12 shows
the photcgraph of a typical Yagi-uda antenna structure.

-0.2 A-&-0.12 A 4

Fig. 7.11
310 MICROWAVES AND RADAR

Fig.7.12
8.10. V-Antenna and Rhombic Antenna
It is a type of long wire antenna. In a V-antenna, the conductors
are arranged to form a V-shape and they are fed in phase opposition at
the apex (Fig. 7.13~).Such an arrangement produces a high gain
bidirectional pattern as shown in Fig. 7.13 (b). If each leg of the
V-antenna is about '8h'long, a 12dB gain over a half wave dipole can
be achieved. The apex angle is very significant in deciding the shape of
the directional pattern. It varies from abut 35" for an '8h' long legs to
about 70" for '2h' long legs. If the antenna is to be used as a wide band
antenna, the apex angle is a compromise between an optimum for the
lowest and the highest frequencies in terms of number of half.

Fig.7.13
ANTENNAS 311

wavelengths in each leg. At the lowest frequencies, the legs should be

several wavelengths long. Such an arrangement producing a bidirec-
tional pattern is a resonant antenna. If this antenna is terminated in a
resistance of about 504 ohms, it becomes a non-resonant antenna and
its directional pattern changes into a uni-directional pattern of the type
shown in Fig. 7.13(c)
In a rhombic Antenna, conductors are arranged to form arhombus.
It is combination of two long wire V-antennas (Fig. 7.13~).In this case
too, the length of the legs and the apex angle control the shape and
directivity of the pattern. The gain of a rhombic antenna, whose
individual legs are of the same lengths as those of a V-antenna, will be
approximately twice. The resonant Rhombic antenna produces a bidi-

(a)
Fig. 7.14

terminating

individualradiation
patterns

Fig.7.15
rectional radiation pattern as shown in Fig. 7.13 (b). If a rhombic
antenna is terminated in a resistance of about 600 to 700 ohms, its
312 MICROWAVES AND RADAR

directional pattern becomes a unidirectional one ( Fig. 7.15 1. This

Rhombic antenna is also called a non-resonant rhombic antenna. The
main features of a rliomlsic antenna are its high gain and wide
bandwidth. These antennas are particularly advantageous at UHF as
the size of antenna legs becomes practicable. An apex angle of 30" is the
most suitsble.

Prvbkm 7.12 :A half wave dipole having a cross-section of 'a'sq.

cm. war; 1br:nd to have an input impedance of 72 ohms. C ~ l c u l a t ethe
inpvt i ~ ~ p e d u n for
c e a folded dipole with the folded portion having an
area of c.-oss-section equal to '1.5a' sq. c n .
Soietiorc : The impedance of a folded dipole is given by the
ex~reseiun:

Zip-=Zi x

= 72
sum of areas of corss-section of various components
Area of cross-section of the driven element
[(a+ 1.5a)/aj2= 72 [2.5a/a12 = 6.25 x 72 = 450 Zl
I'
Problem 7.13 : Design a 3-element yagi antenna suitable for
receiving VHF transmission at 60 MHz.
Solution : h = c / f = 3 x 108/60 x lo6 = 5 rn
Length of dipole = hP2 = 5/2 = 2.5 m
The actual physical length for which the dipole is to be cut should
be about 5% shorter than this calcdated value to compensate for the
increase in length du.e to parasitic effects.
Therefore, length of the dipole = 2.5 - 0.05 x 2.5 = 2.375 m
Length of Director = 2.375 - 0.04 x 2.375 = 2.28 m
Length of Reflector =: 2.375 + 0.05 x 2.375 = 2.49 m
Director-dipole spacing= 0.12h = C.12 x 5 = 0.6 m
Reflector-Dipoie spacing= 0.2h = 0.2 x 5 = 1m

7.11. Reflector Adennas

A Reflector Antenna, made in different types, shapes and con-
figurations depending upon the shape of the reflector and the type of
feed mechanism, is by far the most commonly used antenna type in all
those applications that require high gain and directivity. High gain and
highly directional radiation pattern, which are antenna parameters
that are essentially the same, are the characteristics typical of both
terrestrial and satellite base communication links, radar systems,
direction finding systems and so on. While communicating in the UHF
and microwave frequency bands, it is important to have narrow beam
width to avoidinterference with other transmissions. In a radar system
such as a tracking radar for instance, in addition to the accuracy with
ANTENNAS

which angular co-ordinates are determined, it is equally important to

have a higher angular resolution. Angular resolution, which is the
ability to discriminate between two targets located close to each other,
again depends upon the narrowness of the beam width. Narrower the
beam, higher is the angular'resolution. Now the gain or the directivity
of the antenna is directly proportional to the size of the antenna. The
antenna dimensions need to be much larger than the operating
wavelength for achieving high directivity, a requirement which would
not be practicable at relatively lower frequencies. At UHF and above,
it does become practicable. For example, a t 10 GHz,h = 3 cm and a 3 m
diameter dish would give a dimension that is 100 times the operating
wavelength. Of course, there is small overlap region between VHF (30
to 300 MHz) and UHF (300 to 3000 MHz) and some of the antenna types
to be used for higher end VHF and lower end UHF are common.
A Reflector Antenna in essence comprises of a reflector and a feed
antenna. As mentioned earlier, depending upon the shape of the reflec-
tor and the feed mechanism, there are different types of reflector
antennas suitable for different applications. The reflector is usually a
paraboloid also called parabolic reflector or a section of a paraboloid or
cylindrical. A cylindrical reflsctor has a parabolic surface in one direc-
tion only. The feed mechanisms include the feed antenna placed at the
focal point of the paraboloid oi the feed antenna placed off the focal
point. Another common feed mechanism is the cassegrain feed.
Cylindrical reflectors are fed by an array of feed antennas. The feed
antenna is usually a dipole or a horn. These antennas are thus available
in many types and configurations, some of the more commonly used
ones being :
1. Focal point fed paraboloid reflector (Fig. 7 . 1 6 ~ )
2. Offset fed sectioned paraboloid reflector (Fig. 7.16b)
3. Cassegrain fed reflector (Fig. 7.16~)
4. Array fed cylindrical reflector (Fig. 7.16 d)
Fig. 7.17 shows photographs of some common varities of reflector
antennas.

Feed
314 MICROWAVES AND RADAR

I 1

Fig. 7.17
ANTENNAS 315

Focal Point Fed Paraboloid Reflector

A paraboloid or a parabolic reflector is a three dimensional surface
obtained by a parabola of rotation about its axis and it has a very
important geometric property that for any point on the surface of the
paraboloid, the sum distances of this point from the focal point and the
directrix (shown dotted in Fig. 7.18) is constant. Now, if the source of
radiation is placed at its focal point, the waves travelling towards the
reflector after reflection from different points on the reflector surface
will reach the directrix in phase due to the equal path length distances
involved with the result that the emitted beam is highly concentrated
along the axis of the antenna. Waves originating from elsewhere after
reflection will reach the directrix with phase differences due to path
length differences and get cancelled. Similarly on reception by such an
antenna, the waves approaching the antenna parallel to the axis get
focused onto the feed antenna where as the waves arriving from an
off-axis direction will focus onto different points off the focal point and
thus get diffused. Such a phenomenon makes this type of antenna
inherently a highly directional one. This phenomenon can also be
viewedby drawingthe wave diagram representingpoints of equal phase
in RF energy and as shown in Fig. 7.18. What is shown in Fig. 7.18 is
a ray diagram where rays from focus reflect parallel to the axis and rays
parallel to the axis reilect to focus. In the wave diagram, spherical waves
from focus get reflected and become planar and plane waves after
reflection become spherical.
The basic design parameters of a focal point fed reflector antenna
are the size of reflector, focal length, the feed antenna's beam pattern
and the feed blockage of the reflector aperture.

C 4
! Focal '
length

Fig. 7.18
The size of the reflector and its illumination pattern determine the
antenna gain, the beam width and to some extent the side lobe pattern
too. The power gain of the antenna which in case of as lossless antenna
is same as the directivity is given by :
Ap = 4n ~ e / h ~
316 MICROWAVES AND RADAR

where Ae = Effective aperture area and h = Operating wavelength

,Spherical waves

Fig. 7.19
We know that Ae is less than A due to non-uniform illumination
of the reflector and is related to A by Ae = q. A where 'q' is the aperture
efficiencyor effectiveness. The aperture efficiency varies from 0.6 to 0.7
and is typically 0.65 for a dipole fed reflector antenna. This gives :

Now, A = d2/4where D = Mouth diameter of the reflector

For a dipole feed, q = 0.65 which gives Ap = 6 [ 0/h12

The 3 dB beam width for such an antenna is given by :
70 [LID]
which is a simplified expression but is fairly accurate for reference with
large 'D/h' ratio. The null-to-null beam width is twice the 3 dB beam
width. If the feed beam width is excessive, it causes spill over producing
an undesired antenna response in the direction of spill over. And if it is
too small, only a portion of the reflector is illuminated with the result
that antenna produces a wider beam and a consequent lower gain.

Large focal length Small focal length

Fig. 7.20
ANTENNAS 317

Focal length is another important design parameter, moreso for

tracking radar antennas. In general, longer the focal length, more will
be the displacement for the focus spot of the received energy from the
reflector's focal point for a given angular displacement of the received
wave from the antenna axis. Thus a long focal length reflector antenna
in a tracking radar would produce more error at the feed than that
produced by a short focal length reflector. This is illustrated in Fig. 7.20
The focal length however can not be increased without due con-
sideration of the aperture blockage it causes. Long focal length reflec-
tors need greater support for the feed and hence contribute to a greater
aperture blockage. The directional pattern of the feed antenna deter-
mines the illumination of the reflector. Refer to Fig. 7.21. The angle
subtended by the feed antenna at the edges of the reflector is given

Fig. 7.21
01 = 4 tan-'
[1/(4f/D)]
According to a thumb rule, the 3 dB beam width should be equal
to 0.9 times the subtended angle. According to anothr thumb rule, the
null-to-null bandwidth should equal the subtended angle.
Offset Feed Paraboloid Reflector
Feed together with its supports is one of the major causes of
aperture blockage which is further one of the major causes of side lobes.
In applications where the feed antenna is rather large so as to block a

Fig. 7.22
portion of the reflector aperture with significant effects on the radiated
beam in terms of increased side lobe content, offset feed paraboloid is
318 MICROWAVES AND RADAR

one of the solutions. Fig. 7.22 shows the arrangement of feed and
reflector in this type of antenna. As shown in the figure, the reflected
waves do not get blocked by the feed.

Cassegrain Feed Paraboloid Reflector

Some of the shortcomings of the focal point fed paraboloid reflector
antennas such as aperture blockage particularly when the feed is large
and lack of control over the main reflector illumination can be overcome
by adding a secondary reflector. The contour of the secondary reflector
determines the distribution of the power along the main reflector
thereby giving control over both amplitude and phase in the aperture.
The cassegrain antenna derived from telescope designs is the most
commonly used antenna using multiple reflectors. The feed antenna
which illuminates the second:vy reflector, also called the subreflector,
is a hyperboloid. The reflection from the secondary reflector in turn
illuminates the main or primary reflector. One of the foci of the secon-
dary reflector and the focus of the main reflector coincide and the feed
is placed a t the other focus of the secondary reflector. The arrangement
is shown in Fig. 7.23.

\C-----
Fig. 7.23

Fig. 7.24
ANTENNAS 319

Symmetrical cassegrain systems usually produce large aperture

blockage due to the secondary reflector which can be minirnised by
choosing the diameter of the secondary reflector to be equal to that of
the feed. Blockage can be completely eliminated offsetting both the feed
and the secondary reflector as shown in Fig. 7.24. Such an antenna is
capable of providing very low side lobe level.
Array Fed Cylindrical Reflector
The cylindrical paraboloid antenna as shown in Fig. 7.25 uses a
reflector which is a parabolic surface only in one direction and is not
curved in the other. I t is fed from an array of feed antennas as shown.
This gives a much better control over reflector illumination as compared
to a focal point feed reflector antenna. Electronic steering of the output
beam is also more convenient in an array feed cylindrical antenna.
Symmetrical parabolic cylindrical reflectors suffer from a large aper-
ture blockage. A cylindrical reflector fed from an offset placed multiple
element line source offers excellent performance.

Fig. 7.25

Problem 7.14 : A focal point feed parabolic reflector antenna has

the following characteristics of its reflector :
320 MICROWAVES AND RADAR

Mouth diameter = 2 m
Focal length = 2 m
l i t h e 3 d B beam width of the antenna has been chosen to be 90%
of the angle subtended by the feed a t . he edges of the reflector, determine
t h e 3 d B beam width and the null-to-null beam width ofthe feedantenna.
Solution :The angle subtended by the focal point feed at the edges
of the reflector '81' is given by :
Bsub = 4 tan-' [1/4v / D ) ]
wnere f = focal ler.gti1
D = mouth diameter
Bsub = 4 tan [ 1 / ( 4x 2/2)] =4 tan 0.25 = 56"
Therefore, 3 dB beam width = 0.9 x 56' = 50.4"
Null-to-null beam width = 2 x (3 i B beam width) = 100.8".
Problem 7.15 : A symmetrical cassegrain reflector antenna uses
a main paraboloid reflector with a focc.2 length of 3 m. If the feed is placed
at a distance of 1.5 m from the point of intersection of the secondary
reflector and the antenna axis, determiae the focal length of the hyper-
boloid secondary reflector.
Soluti~n :Refer to Fig. 7.26.I'he feed is placed at one of the foci
of the hyperboloid and the other fceus of the same is coincidentwith the
focus of the paraboloid.
The focal length of the hyperboloid from the figure = 3 - 1.5 =
1.5 m.

Fig. 7.26

Problem 7.16. A paraboloid reflector antenna uses a focal point

feed having a 3 d B beam width of 63".If the mouth diameter of the
reflector is 3m, determine the distance of the feed from the point of
intersection of antenna axis and the reflector surface assuming that
beam width is 0.9 times the subtended angle.
Solution. The subtended angle Bsub' is given by
Bsub = 4 t a n 1 [1/(4f/D)]
Now, 0sub = 63/0.9 = 70"
ANTENNAS 321

Therefore, 4 tan-' [1/(4f/3)] = 70" or tan-' [1/1.333 fl = 17.5"

This gives, l/l.333f = tan 17.5"= 0.3152
or f= 1A.333 x 0.3152 = 2.38 m

8.12. Lens Antenna

Like reflector antennas such as paraboloid reflectors, Lens Anten-
nas are another example of application of rules of optics to microwave
antennas. While in the case of former, it is the laws of reflection, the
lens antennas depend for their operation on the refraction phenomenon.
Lens antennas are made of dielectric material. Fig. 7.27 (a)illustrates
the principle of operation of such an antenna. A point source of radiation
is placed at the focal point of the lens. The rays arriving at the lens
closer to the edges of the lens encounter a larger curvature as compared
to those arriving at the center portion of the lens. The result is that the
rays close to the edges are refracted more than the rays closer to the
center. This explains the rays getting collimated and becoming parallel
to the lens axis after passing through the lens though they are inclined
in the space between the lens and the point source. Similarly, on
reception, the rays arriving parallel to the lens axis get focused on to
the focal point where the feed antenna is placed.

(a) (b)
Fig. 7.27
Another way of explaining the same is to look at the wave diegram
as shown in Fig. 7.27(b).Spherical waves emitted by the point source
get transformed into plane waves during transmission. The reason for
this is that those portions of the wavefront closer to the center get slowed
down relatively more than those portions that are closer to the edges.
With the result that the outgoingwaves are planar. By same reasoning,
plane waves incident on the lens antenna during reception emerge as
spherical waves travelling towards the feed.
Now the precision with which these transformations take place
depend upon the thickness of the lens in terms of the operating
wavelength. In fact the thickness of the lens at the center should be
much larger than the operating wavelength. This makes lens antennas
less attractive at lower microwave frequencies. As an example, for an
operating frequency of 3GHz, h = 10 cm. Even a'1OX thickness requires
322 MICROWAVES AND RADAR

a lens that is 1 m thick at the center. It is because of this reason that

the lens antennas are not the favoured ones for frequencies lower than
10 GHz. Even at frequencies around 10 GHz too, the problems of
thickness and weight are there. These can be overcome by using what
is called a Fresnel lens or zoned lens. Two types of zoned lenses are
shown in Fig. 7.28. Zoningnot only overcomes weight problem, it would
also absorb less energy. A thicker lens would absorb a higher proportion
of the radiation. The thickness 't' of each step in a zoned lens is related
to the wavelength of the signal to ensure that phase difference between
the rays passing through the center and those passing through the
adjacent section is 2n radians or an integral multiple of it. A zoned lens,
because of the step thickness being related to the operatingwavelength
has a small operational frequency range.

Fig. 7.28
8.13. Horn Antenna
Just as in case of a transmission line open circuited at the load
end,not all the electromagnetic energy is reflected and some does escape
to the surrounding atmosphere, the same is true for waveguides too.
This radiation of electromagnetic energy is however inefficient due to
a combination of factors, the most prominent being the impedance
mismatch between the transmission line (or the waveguide) and the
atmosphere. While explaining the radiation mechanism of an antenna
in the earlier part of the chapter, it was mentioned that the energy
coupling to the atmosphere could be enhanced and radiation efficiency
significantly improved by opening out the open end of the line and
straightening the conductors so as to take the shape of the dipole. If the
same principles are applied in case of waveguides to improve the
coupling of electromagnetic energy to the atmosphere, the waveguide's
abrupt discontinuity is tfansformed into a more gradual one. What we
get is a Horn-Antenna ahd we get a Sectoral Horn (Fig. 7.29a) where
the flare is only on one side, Rectangular Pyramidal Horn (Fig. 7.293)
where the flare is on both sides and the ConicalHorn (Fig. 7.29~) which
is a natural extension of a circular waveguide.
The important design parameters of a horn antenna include flare
length and flare angle (Fig. 7.29d). The flare angle can not be either too
small or too large. If too small, the antenna has low directivity and also
ANTENNAS 323

the emitted waves are spherical and not planar. Too large a flare angle
also leads to loss of directivity due to diffraction effects.

Fig. 7.29
Horns could have simple straight flares or exponential flares.
These are used very commonly as the feed antennas for the reflector
type antennas. When more demanding antenna performance is desired
in terms of polarisation diversity, low side lobe level, high radiation
efficiency etc., the feeds also become more complex. Segmented (Fig.
7.30a), finned (Fig. 7.30b) and multimode (Fig. 7.30~)horns may be
used.

Fig. 7.30
324 MICROWAVES AND RADAR

Some combinationofhorn antennas :.nd paraboloid reflectors such

as Cass-Horn and Hcg-Horn antennas have gain and 5eam width
specifications matching those of paraboloid reflectors of comparable
dimensions.
7.14. Helical Antenna
Helical antenna is a broadband VHF and UHF antenna and in
addition to its broadband capabilities, it has found most of its applica-
tions due to the circularly polarised waves i t produces. VHF and UHF
propagation undergoes a random change in its polarisation. as it
propagates through the atmosphere due to various factors like earth's
magnetic field, ionisation of different regions of the atmosphere and so
on, Faraday's rotation being the main cause. The propagation gets more
severely affected in case of trans-ionospheric communications such as
those involving satellites. Circular polarisation is to a great extent
immune to these polarisation changes. On the other hand, a horizon-
tally polarised wave will not be received at all if its polarisation got
rotated by 90" and became vertically polarised.
Ground Plane
,Helix

Fig. 7.31
Fig. 7.31 shows a typical helical antenna. The ground plane is a
wire mesh. This antenna has two modes of operation with one producing
circularly or elliptically polarised broadside pattern with the emitted
wave perpendicular to the helical axis and the other producing a
circularly polarised endfire pattern with the emitted wave along the
helical axis. For the first mode, the helix circumference is much smaller
than the operating wavelength whereas for the second mode, which is
the more common of the two, the helix circumference is approximately
equal to the operating wavelength.
7.15. Log Periodic Antenna
Log periodic antenna is another broadband VHF and UHF anten-
na capable of providing enormous bandwidth. It is a driven array and
is made in a very large variety of shapes and configurations. One of the
most commonly used type is a driven array of dipoles as shown in Fig.
7.32(a). The array is driven by a feeder line that is transposed between
adjacent elements so that the feed to a given element is 180"out ofphase
with that to the adjacent elements. The lengths of the dipoles and the
inter-dipole spacing is governed by the following relation.
R1/R2 = R2/R3 = R3/R4 .........
= Li/L2 = L2/L3 = L3/L4 ......... = k (constant)
ANTENNAS 325

Also, the typical values for the convergence angle and constant 'k'
are 30" and 0.7. The lowest and highest frequency of operation are
respectively determined by the longest and the shortest dipoles. The
cut-off frequencies are the ones for which the length is W 2 ' . Straight
dipoles are usually used for the UHF band and the dipoles are bent like
V-antennas as shown in Fig. 7.32(6) for operation in VHF band.

Fig. 7.32
8.16. Phased Array Antenna
A Phased Array Antenna or more appropriately a phase steered
array antenna is the one where the radiated beam (or the axis of the
main lobe of the radiated beam) can be steered by feeding the elements
of the array with signals having a certain fixed phase difference between
adjacent elements of the array during transmission. On reception, they
work exactly the same way and instead of splitting the signals among
elements, the elemental signals are summed. The receive steering uses
the same phase angles as transmit steering from antenna reciprocity
principle.
Phased array antennas find extensive applications in different
types of radars including those used for surveillance, tracking, air
defence and so on. Fig. 7.33 shows photographs of some of these radar
types alongwith their phased array antennas. In fact, a radar using a
phased array antenn is also called a Phased Array Radar.
The elements used in the array are usually either horns or
microstrip antennas and the array can have any one of the large number
of available configurations. A Linear Array is a one dimensional array
with multiple elements along its length with all the elements lying on
a single line (Fig. 7.34~).The type of array would be capable of steering
the beam in only one dimension depending upon the orientation of the
326 MICROWAVES AND RADAR

Fig. 7.33.(Contd....)
328 MICROWAVES AND RADAR

array. A Planar Array is a two dimensional array with multiple ele-

ments in both dimensicns with all elements lying on the same plane
(Fig. 7.3413). The arrays are capable of steering the beam in two
dimensions, azimuth and elevation, for instance. An array is said to be
periodic where the element spacing is constant over the entire antenna
arrav and randam where the element spacing is a periodic. There are
arrays where some elements are not filled or are inactive. Such an array
which is usually periodic is called a Thinned Array.

(a) (6)
Fig.7.34
The basic principle of operation of a phase steered array can be
explained with the help of the array shown in Fig.7.31. Assuming that

Fig. 7.35
the observation point is in the far field of the antenna so that it can be
considered equidistant from various elements of the array along the
array axis. We can then write the following expression :
Ar=SsinO
where 8 = Angle between the observation axis and the array axis
S = Spacing between adjacent elements
Ar = Difference in distance between the observation point and
the two adjacent points
If 'A' is the operating wavelength, a path length diff'erence of 'b'
corresponds to a phase difference of [2n/hl. Ar which gives :
[2n/h.] . Ar = S sin 8. [2x/h] = A$
where A+ = Phase difference between adjacent elements
or A@= [27~S/h]sin 8 .......................................in radian
ANTENNAS 329

A$ = [2nS/hI sin 0 .......................................in radian

or
- [360S/h]. sin 0 .................................... in degrees
Also, sin 0 = h . A$/2xS
or 0 = sin-' [h . A$/2xS]
These expressions can be used to determine the steering angle '0'
for a given adjacent element phase shift and vice versa. As is clear from
the description, the stability of the phase of the signals applied to
different elements of the array is very important. The phenomenor of
beam steering by having different elements fed with signals having
fixed known phase differences can also be explained qualitatively by
considering the fact that in the far field the emitted waves must be all
in phase at the point of maximum gain. With reference to Fig. 7.34, that
would be possible only if the element-:! was fed with a signal lagging in
phase corresponding to distance 'Ar' as it has to travel that much
distance less to the observation point. Similarly, the element-3 should
lag in phase from the signal fed to element-2 by the same amount and
so on. This also explains why the beam is steered in a direction away
from leading phase angles and towards the lagging phase angles. To
sum up, magnitude and direction of steeringrespectively depends upon
the magnitude of the inter-element phase difference and the direction
in which the phase is lagging. Any phase noise in the signals fed to the
elements leads to increased Se'tm width, reduced gain and increased
side lobe level. Also, the beam width of an off-axis beam is slightly wider
than the beam width of the on-axis beam as the effective aperture of
the array would be smaller when viewed off-axis than when it is viewed
on-axis.
Based on the feeding methodology, these electronically steered
arrays are configured in one of the following ways :
1. Conventional Arrays
2. Lens Arrays
3. Reflector Arrays
Conventional Arrays have their elements fed from a common
source by using power dividers and combiners as shown in Fig. 7.36.

Fig. 7.36
The divider 1 combiner can further have any of the configurations
shown in Fig. 7.37.
330 MICROWAVES AND RADAR

End Feed Center Feed

I Equal path 1 Corporate

length feed feed

Fig. 7.37
Lens Arrays have their elements fed from a common source which
radiates spherical waves to one side of elements during transmission.
These elements receive, process and re-radiate a plane wave from the
other side (Fig. 7.38).

A I A z (Azimuth)

I A
( Elevation) E

Fig. 7.38

Short
circuits

Fig. 7.39
ANTENNAS 33 1

Reflector Array is also a space fed array. In this case, elements

receive, process and re-radiate on the same side (Fig. 7.39).
In practice, the phase shifters are either digital or digitally con-
trolled with the result that only discrete phases are possible. For
example, in a 4-bit phase shifter, the phase can have any of the 16
possible values so that the possible phase angles can be On ,22.5", 45',
67.5", 90°, 112"., 135" , 157.5", 18O0,202.5", 245", 247.5", 270°, 292.5",
315", 337.5", and 360". In such a case, the phase of the signal to a given
element would be the one that is closest to the theoretically computed
phase angle for the desired steering angle. A larger bit phase shifter
wouldgive lesser phase error. Ferrite phase shifters and the diode phase
shifters are the commonly used ones.

Problem 7.17 : A linear periodic array of 5 elements has an

inter-element spacing of 10 cm. If the operating frequency is 2.5 GHz,
determine the desiredphases of all elements if the beam is to be steered
by l o o towards the right side of the array axis as shown i n Fig. 7.40.
Phase for element-1 can be taken as zero.

Fig. 7.40
Solution : A$ = [360S/h] sin 0
Now, h = c/f =3x 10/2.5 x 10' = 0.12 m
Therefore, A$ = [360 x 0.1/0.12]. sin 10"= 51".
Therefore, the phase angles for elements 1, 2, 3, 4 and 5 are
respectively On,51°, 102', 153' and 204".
Problem 7.18: Inproblem 7.17, ifthe beam is to be steered towards
the left of the axis with all other parameters remaining the same,
determine the phase angles for various elements.
Solution. The beam steers towards lagging phase angles, there-
fore, the phase angles for elements 1, 2, 3, 4 and 5 would respectively
332 MICROWAVES AND RADAR

Problem 7.19: Fig. 7.41 shows a n array of antenna elements. For

the given phase angle distribution among the different elements, deter-
mine the beam position i n azimuth and elevation for a n inter-element
spacing of 5 c m and an operating wavelength of 6 cm.

Fig.7.41
Solution :From Fig. 7.41,A+ in azimuth direction = 25"
A$ in elevation direction = 35"
Now, the steering angle '8' in azimuth
= sin-' [A. A$/360S]
= sin-' [6x 25/360 x 5 I = sin-' [1/121=4.8"
Similarly, '0' in elevation
= sin-' [ 6 x 35/360 x 51 = sin-' [7/601= 6.7"
Considering the antenna axis to be perpendicular to the plane of
the paper, the azimuth steering is towards the right and the elevation
steering is upwards.

Problem 7.20 : In case of problem 7.19, the phase angles shown

are implemented by using 4-bit phase shifters, determine the actual
phase angles for the signals applied to various antenna elements.
Solution : A 4-bit phase shifter can not resolve phase angles
better than 22.5".The actual phase angle for a given element in the
array will be the one that is one of the resolvable levels between 0"and
360" and also is the closest to the desired value to be implemented. The
actual phase angles'that can be implemented using a 4-bit phase shifter
for the angles shown in Fig. 7.41 are shown in Fig. 7.42.
ANTENNAS

Fig. 7.42

8.17.Microstrip Antenna
Microstrip as a miniature transmission line developed earlier
mainly to take advantage of semiconductor microwave devices and
Microwave Integrated Circuits (MICs) was discussed in an earlier
chapter. It consists of a thin strip sitting on a dielectric that rests on a
ground plane. Microstrip when used as a transmission line was men-
tioned earlier to have a tendency to radiate from irregularities and
sharp corners which indicates that such a component could possibly be
used as an antenna. Microstri~antenna in the form of an arrav is being -
increasingly usedin radar antennas such as phased array antennas due
to its ease of construction and low cost.
Microstrip antennas radiate efficiently as devices on microstrip
printed circuit boards and the microstrip antenna arrays would consist
of microstrip elements, feed mechanism, phasing networks and any
other microstrip devices. The most commonly used microstrip antenna
element is a rectangular element photoetched from one of the sides of
a double sided PCB with the other side used as a ground plane as shown
in Fig. 7.43(a).The element is fed from a co-axial feed. The length 'L'
here is the most critical device dimension and is slightly less than half
the operating wavelength in the dielectric substrate material.
L = 0.49k = 0 . 4 9 h 6
where E~ = Relative dielectric strength of printed circuit substrate
. material
The thickness 't' is of the order of 0.01h. The selected value of 't' is
based on the desired bandwidth and conm~erciallyavailable thickness.
The width 'W must be less than a'wavelengthin the dielectric substrate
material so that higher order modes are not excited. However, this is
334 MICROWAVES AND RADAR

not the constraint if multiple feeds are used to eliminate higher order
modes. Width 'V decides the input impedance of the antenna element.
The expected bandwidth can be conlputed from :
B W = 127. f.t.
where f = operating frequency in GHz
t = thickness in inches
BW = Bandwidth in MHz
The input impedance of the element can be computed from :
Rln= 60 hlj/W for [h/2] element
= 120 ho/W for [h/4] element
The above expressions however assume that the two slots of the
microstrip element are uniformly excited.
Most of the microstrip elements are fed from a co-axial connector
which is soldered to back of the ground plane with the feed pin soldered
to the microstrip element as shown in Fig. 7.43(b).Another microstrip
antenna element is W 4 ' element. It consists of a h/4 long photoetched
element etched in substrate material. It is used for broader E-plane
beam width. A third element is 'h' long. It is similar to W2' element
except that feed is in the center.

Fiq. 7.43
One of the biggest advantage.! of microstrip antenna arrays is that
impedance matching, power division and power combination are very
easy to implement. These features make it very attractive to be used in
electronically steered array antenna. Fig. 7.44 shows, as an example,
an array of four elements. Assuming that the feed point is 50Cl and the
antenna element is 150Q the power divisions are as shown. The 71R
and 122R line sections are each 'h/4' long. The first provides a match
between lOOR and 50R sections and the second does it for lOOQ and
150R sections. We would recall that a W 4 ' line section has a property
that it provides a match between impedances Z1 and Zz provided. its
characteristic impedance is equal to m.
Another significant advantage of microstrip antenna is the ease
with which it can produce circular polarisation. Fig. 7.45 shows
ANTENNAS 335
microstrip realisation of this. In this case, the elament is fed from two
points, one producing horizontal polarisation due to feed at the left and
the other producing vertical polarisation due to feed at the bottom. The
feed track is offset so that the signals fed to the element are having a
phase difference of 90" m e result is a circularly polarised wave
radiated normal to the plane of horizontal and vertical polarisations
and perpendicular to the plane of paper. The impedance matching line
sections are not included here for the sake of simplicity.

Fig. 7.44

Microstrip
- +45" Element

-45" 3
50 n
Feed Point
loon

Fig. 7.45
336 MICROWAVES AND RADAR

Review Questions
I. What functions are performed by an antenna ?
What do you understand by Antenna Reciprocity ?
2. Differentiatebetween a resmant and a non-resonant antenna
with particular refxence to directional pattern.
3. Briefly describe the following antenna parameters :
(a) Directivity
( b ) Polarisation
(c) Antenna Aperture
( d ) Input Impedance
4. Compare the characteristics of a half wave dipole and a
Marconi antenna.
5. Compare the characteristics of a half wave folded dipole and
a three element Yagi antenna.
6. Briefly describe the operational principle of a parabolic reflec-
tor antenna. What are the advantages and limitations of using
a cassegrain feed in a refledor antenna ?
7. Which factor limits the use of lens antennas even at microwave
frequencies ? How does zonicg help to overcome some of the
shortcomings ?
8. Briefly describe the operational principle of a phased array
antenna. Where does such antenna find main application ?
9. Name any three broad9ar.d antenna types. Compare their
performance with particular reference to their percentage
bandwidth.
10. Write short notes on :
(a) Microstrip antenna
( b ) Antenna's physical and electrical lengths
(c) Broadside and Endfire arrays

Problems
1. Determine the directivity of an antenna having a radiation
efficiency of 80%and a power gain of 40.
[17dB]
2. An antenna has directivity and power gain specificatiocsof 17
dB and 30 respectively. If its loss resistance is 6l2, determine
the radiation resistance.
19 n1
3. Determine the power gain of an antenna having an aperture
area of 10 m2 and an operating frequency of 10 GHz.
[51.45 dB]
ANTENNAS 337

4. Received signal strength in a certain vertically polarised an-

tenna is 30 dB when it is receiving a right hand circularly
polarised wave. Compute the received signal strength when
(a) incident wave is vertically polarised
(b) horizontally polarised ,
(c) left hand circularly polarised
[(a) 33 dB, (b) 0 dB and (c) 30 dB]
5. A focal point fed paraboloid reflector antenna having a feed
antenna beam width of 70" fills only 90% of the reflectois
mouth diameter which is 4 m. Compute the distance of the feed
antenna from the point of intersectioinof antenna axis and the
reflector surface.
[2.832 ml

SELF EVALUATION EXERCISE

Multichoice Questions
1. The resistance value of a hypothetical resistor that would
dissipate a power equal to the power radiated by the antenna
when fed with the same current is known as
(a) intrinsic impedance of the antenna
(b) equivalent noise resistance of the antenna
(c) loss resistance of the antenna
(d) radiation resistance of the antenna
2. Directivity of an antenna is
(a) same as its directive gain
(b) maximum value of its directive gain
(c) directive gain expressed in decibels
(d) directly proportional to its beam width
3. Antenna reciprocity is put to use in
(a) radar systems
(b) television broadcast
(c) commercial radio broadcast
(d) EW systems
4. For an isotropic radiator, the radiation intensity for a total
radiated power of 'P'watts is given by :
(a) P/2n watts per steradian
(b) P/4n watts per steradian
(c) 4 e watts per steradian
(d) none of these
338 MICROWAVES AND RADAR

5. The antenna of Fig. 7.46(a)will have a radiation resistance of

( a ) 7 3 ohms ( b ) 36.5 ohms ( c ) 146 ohms ( d ) 292 ohms
(Asssume the antenna to be lossless)

(b)
Fig, 7.46
6 . The antenna of Fig. 7.46 (b)will have a radiation resistance of
( a ) 7 3 ohms ( b ) 36.5 ohms
( c ) 292 ohms ( d ) 146 ohms
(Assume the antenna to be lossless)
7. The antenna shown in Fig. 7.46(b)
(a) is a Yagi-Uda antenna
( b ) is a Hertz antenna
( c ) is a Marconi antenna
( d ) produces horizontally polarised waves
8. One of the following is a non-resonant antenna
( a ) Rhombic antenna
( b ) Hertz antenna
( c ) Marconi antenna
( d ) Yagi antenna
9. One of the following antenna types has a very large bandwidth
(a) Yagi antenna
( b ) Parabolic reflector
(c) Log periodic array
( d ) Hertz antenna
10. A helical antenna produces
( a ) horizontally polarised waves
(b) vertically polarised waves
( c ) circularly polarised waves
( d ) linearly polarised waves
11. A resonant antenna is characterised by
ANTENNAS 339

( a ) presence of standing waves

( b ) a unidirectional radiation pattern of high directional gain
( c ) a bidirectional radiation pattern
( d ) both ( a )and (c)
12. In a linear phased array antenna, the incremental phase shift
required to produce a steering angle of 30"for an inter-element
spacing of 10 cm and an operating wavelength of 20 cm would
be
( a ) 90" ( b ) 45" ( c ) 30° ( d ) 180"
13. In a phased array antenna, the beam steering is always
( a ) towards leading phase angles
( b ) towards lagging phase angles
( c ) in one dimension
( d ) three dimensional
14. A digitally controlled 4-bit phase shifter in a phased array
antenna will implement a phase angle of 150" by (assuming
a 0" reference phase)
( a ) 157.5" ( b ) 150" ( c ) 135O (d) 152.5"
15. In a parabolic reflector antenna, the electromagnetic waves
between the feed and the reflector are
( a ) spherical waves
( b ) planar waves
( c ) highly directional
( d ) omni-directional
 MICROWAVE
MEASUREMENTS
AND INSTRUMENTATION

n this chapter, we shall discuss different types of measurements that

I are usually carried out with microwave devices and circuits. The test
equipment that form a part of the measurement set up in carrying
out certain specific measurements shall also be discussed briefly as
regards their operational fundamentals. Different microwave measure-
ments to be covered in the present chapter mainly include Transmission
and Reflection measurements (including measurement of scattering
parameters), measurement of power, frequency, noise and impedance.
Commonly used test equipment include Microwave Frequency Counter,
Spectrum Analyser, Network Analyser and so on.
The topics covered in this chapter include :
Scattering Para- 8.7. Microwave Power
meters Measurement
Transmission and 8.8. Noise Measure-
Reflection ment
Parameters 8.9. Phase Noise
Network Analyser 8.10. Time Domain
Measurement Un- Reflectometry
certainty 8.11. Spectrum
Measuring with Analysers
Scalar Network 8.12. Frequency
Analysers Counters
Measuring with 8.13. Frequency Syn-
Vector Network theslsers and Syn-
Analysers thesised Functlon
Generators
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 34 1

8.1. Scattering Parameters

Scattering Parameters can be used very convenientlyfor determin-
ing signal flow conditions when a device is inserted in a microwave
circuit. The scattering parameters of this device, called the Device
Under Test (DUT),can then be used to determine various transmission
and reflection parameters. Fig. 8.1 shows a DUT as a two port network.
Here a l , a 2 are the signals entering the device while blm b2 are the
signals leaving the device. The signal flow conditions can then be
written in the form of two linear equations involving the scattering
parameters sll, s12, s21, s 2 2 andal, a2, bl, b2 as :
bi = s i i . a1 + s12 . a 2 (1)

Fig. 8.1
If there is a 'n' port network, then the set of 'n' linear equations
characterising the signal flow conditions in that network could be
written as :

These equations can also be represented in the matrix form as :

The above representation can be written in the short form as [b] =

[s] . [a] where [s] is called the Scattering Matrix. The expressions for
various scattering parameters can be determined by a two step process.
In the first step, output of the DUT (Device Under Test) is terminated
with an impedance equal to the characteristic impedance '20' as shown
in Fig. 8.2, then 'a2' becomes zero. Substitutingaa = 0 in equations (1)
and (21, we get :
bi = s i i . ai, whichgives
342 MICROWAVES AND RADAR

sii = (bilai), a 2 = 0

Fig. 8.2.
Thus 'sll' can be defined as the input reflection co-efficient when
output is terminated in '20'. Substituting a2 = 0 in equation (2), we
get:
b2 = sai . a1 which gives
s 2 1 = (bz/al), a 2 = 0
' ~ 2 1 'can be defined as the forward transmission gain (or Insertion
Gain of the DUT) with output terminated in '20'.
In the second step, input is terminated with '20, as shown in Fig.
8.3. This implies that a1 = 0. Substituting a1 = 0 in equations (1)and
(21, we get
bi = a12. a2 which gives

' ~ 1 2 'is the reverse transmission co-efficientwith input terminated

in '20'.
Also, bz = s22 . a2 which gives
s22 = (b2/a2), a1 = 0
' ~ 2 2 'is the output reflection co-efficient with input terminated in '20'

Fig. 9.3
Scattering parameters offer very simple relationship between
them and the various power signals a t the input and output of the device
or network under test. Following expressions are useful.
1 sll 1 = (Powerreflected from network input/Power incident on
network input)
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 343

( s22 l 2 = (Power reflected from network output/Power incident

on network input)
I s2 1 1 = (Power delivered to Zo load1Power available from Zo
source).
$21' is the forward transmission with Zo load.
( s12 ( = (Power reflected from input port of the network / Power
reflected from the load)
's12' is the reverse transmission with Zo source and Zo load.

8.2. Transmission and Reflection Parameters

At microwave frequencies, the operational wavelength becomes
comparable to the dimensions of the device geometries. These com-
ponents actually take the form of circuits consisting of the basic ele-
ments and the parasitics like lead inductances, stray capacitances and
so on. At microwave frequencies, it is more appropriate to consider it
like a wave and study how it responds to a component in its path as it
propagates through a transmission medium, an approach similar to the
one followed in case of light waves. The components, at microwave
frequencies, can thus be characterised by various transmission and
reflection parameters. As we shall see, some of the parameters are
represented by the S-parameters discussed above.
Reflection Parameters
The major reflection parameters include :
1. Reflection Coefficient, r
2. Standing Wave Ratio (SWR)
3. Return Loss
Reflection Coefficient :The Reflection Coefficient is defined as
(Refer to Fig. 8.4)
V = Vref/Vinc = (ZL- ZO)/(ZL+ 20)

u Fig.8.4
where r = Reflection coefficient
2 0 = Characteristic impedance of the
transmission medium
ZL= Load impedance or in the present case the input impedance
of the device under test (DUT)
344 MICROWAVES AND RADAR

Vim = Incident wave

V,,f = Reflected wave
The reflection coefficient 'T'is a vector quantity and therefore has
both magnitude as well as phase. The magnitude of the reflection
coefficient is represented by 'p' i.e. p = I r 1 .
Standing Wave Ratio (SWR):The SWR is defined as :
SWR = (1+ p)/(l - p)
In w e , 'ZL'is a complex load impedance,
SWR = ZL/ZO= (1+ r)/(l- l?)
-urn Loss :Return loss is expressed in decibels by :
Return Loss (in dB) = - 20 log p
Thus Return Loss varies from 0 dR to infinity for magnitude of
r e k t i o n coefficient varying from '1'(100% reflection) to '0' (for zero
reflectim). As reflection parameters such as 'I" , SWR originate from
impedance mismatches as is clear from the expressions representing
these parameters, measurement of these parameters can be used to
study the impedance characteristics of the device :
Transmission Parameters
Important transmission parameters include
1. Transmission Coefficient
2. Insertion Loss 1 Gain
Transmission Coefficient :Transmission coeficient is defined
as (Vbam/Vim) where 'Vim7 and V ' t,' represent the incident and
transmitted waves through the DUT.
Insertion Loss :Insertion Loss is expressed by :
Insertion LOSS(in dB) = 20 log [ IVt,, ( I/[ I Vine ( I
Insertion Phase : It is the difference between relative phase
angle of the transmitted wave and that of incident wave.
Insertion phase = L Vtm, -V !,m
i
8.3. Network Analyser
Network analyser is the key instrument used for making a variety
of measurements on passive as well as active microwave components
and networks. Network analysers can measure both impedance (Reflec-
tion) and gain (~ransmissibn)characteristics of the devices. Since
networks can be completely characterised by impedance and transmis-
sion characteristics, it is an indispensable instrument for optimising
the microwave circuit design. It characterises both active and passive
networks by creating a data model of component parameters like
impedance and gain characteristics through stimulus-response testing
over the fiequencyrange of interest. The parameters vary as a function
of hquency and have complex variables of both magnitude and phase.
Swept network analyser can be used to measure magnitude and phase
as a function of frequency. Network analysis is generally limited to
characterisation of linear networks. For a linear network, a sine wave
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 345

stimulus p~oducesa sine wave output. Sine wave testing is therefore

an ideal method for characterising magnitude and phase response as a
function of frbquency.
A network analyser that measures complex reflection and trans-
mission characteristics of microwave devices as a function of frequency
is the Vector ?Jetwork Analyser as compared to a Scalar Analyser that
provides only the magnitude characteristics. Both vector as well as
scalar network analysers device the desired information by cornpaling
the incident wave with the transmitted or reflected waves depending
upon the parameter to be measured. The scattering matrix of the DUT
is derived by taking the ratio of relevant signals. Each S-parameter is
a complex quantity having magnitude and phase information. these
parameters represent device's input reflection coefficient 'Sii',forward
transmission 'S219,reverse transmission '512' and output reflection
coefficient 'S22' (Fig. 8.5).

Fig. 8.5
Fig. 8.6 shows the basic block schematic arrangement of a network
analyser. It has four major components namely :
1. Signal source
2. Signal separation device
3. Receiver
4. Processor/Display

Transmitted
Reflected

I Signal Separation I

u ~ e c e i v e rDetector
l

e Processor/ Display

Fig. 8.6.
346 MICROWAVES AND RADAR

The Signal Source provides the source of incident or stimulating

signal to which the DUT responds by transmitting a part of it and
reflecting the remaining part. The source frequency can be swept to plot
magnitude versus frequency and phase versus frequency charac-
teristics. Signal sources are either conventional sweep oscillators or
synthesised sweepers including the synthesised signal generators.
While the sweep oscillator offers a low cost solution, it heavily suffers
in terms of frequency accuracy and stability. Frequency stability is very
important particularly when the DUT phase characteristics varyrapid-
ly with frequency. Frequency range, signal purity, frequency stability
and output power level are the important specifications of a signal
source to be used in a network analyser.
After the DUT is stimulated by the signal, the device responds by
partially transmitting and partially reflecting the same. The next
obvious step is to separate the incident, transmitted and reflected
signals so as to be able to measure magnitude and phase of individual
components. The devices commonly used for signal separation are the
Directional Coupler, Power Splitter and the High Impedance Probes.
Either of these can be used for the purpose. Fig. 8.7 shows the use of
directional coupler in a transmission measurement environment.Direc-
tional coupler, as we know, consists of coupled transmission lines
configured in such a way that a fraction of energy passing through the
main port in a particular direction gets coupled to the auxiliary port.

Incident
I m
-
I

Receiver Display

DUT
Source D~rect~onal
Coupler Transm~tted
Fig. 8.7
Fig. 8.8 shows the use of a power splitter (two resistor power
splitter) for the same purpose. The input signal here is split equally
between two arms with the output signal from each arm being 6 dB
below the input signal level. Power splitter provides a very good source
match. If one of the split outputs is considered as a reference and the

Fig. 8.8
MICROWAVE MEASUREMENTSAND INSTRUMENTATION 347

other goes to the DUT,then ratio of transmitted to incident signals has

the effect of making the resistor in the power splitter determine
equivalent source match of measurement. Also, power splitters are
broad band and offer excellent frequency response.
High impedance probes (Fig.8.9) can be used for making measure-
ments in environments different from 50 or 75 ohms
Signal separation in case of reflection measurements can be
achieved by using either a dual directional coupler or a bridge. A bridge
is broad band while the directional coupler has lesser main arm loss.

Fig. 8.9
The Receiver down converts the microwave signal to a DC or a
lower IF. The commonly used receiver techniques used include :
(a) use of a diode detector as a broad band sensor which converts
microwave input to a DC or an audio frequency square wave
with its magnitude proportional to the incident power as
shown in Fig. 8.10.

(b) tuned receiver using mixing at fundamental component as

shown in Fig. 8.11.

Fig. 8.11
(c) tuned receivers using harmonic mixing as shown in Fig. 8.12.
348 MICROWAVES AND RADAR

The tuned receivers have a narrow band pass IF filter to reject

spurious signals and extend the noise floor of the receiver. While broad
band diode detection is used in scalar network analyser, tunedreceivers
are used in vector network analyser.
The detected signals are then processed and displayed. Network
analysers are multi-channel receivers utilising a reference channel and
at least one test channel. Different parameters that can be displayed
include the absolute signal level in hfferent channels, ratio of signal
levels between different channels, relative phase difference between the
channels etc. Relative ratio measurements are usually made in dB.

Harmonic
Generator I

Fig.8.12
8.4. Measurement Uncertainty
The accuracy of measurement made .with a network analyser
depends upon various sources of error or uncertainty related to the
network analyser, the measurement technique employed and even the
device under test. There are three broad categories of errors that affect
measurement accuracy. These are :
1. Systematic errors
2. Random errors
3. Drift and Stability errors.
The overall uncertainty in magnitude can then be expressed as :
Magnitude Uncertainty =
Error)' + Drift Error + Stability Error12
Systematic + d ( ~ a n d o m
Error
Systematic Errors are those errors that do not change after a
calibration has been done and remain constant during the measure-
ment. They are an indication ofhow well the actual measurement set-up
simulates an ideal stimulus-response test environment. For instance,
if the test system source is not perfectly matched to the transmission
medium, the reflected wave from the DUT will be re-reflected of the
non-ideal source andreturn towards the DUT as a second incident wave.
Multiple reflections between source and DUT cause a measurement
error whose magnitude would depend upon the level of mismatch.
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 349

Dynamic accuracy, which is the ability of the detector to accurately

detect a signal over a large amplitude range, is another type of sys-
tematic error.
Random Errors, as the name suggests, are random in nature and
do not remain constant during a measurement. Key contributors to
random errors are noise sources, connector repeatability and cable
stability.
Drift and Stability Erromare due to the changes that occur inthe
system with time and temperahre.
8.5. Measuring W i t h !Scalar Network Analyser
Scalar techniques measwe transmitted and reflected signal mag-
nitude. Scalar network analyser (SNA) can be used to charactwise
devices such as amplifiers, antennas, attenuators, adapters, RF
bridges, duplexers, couplers, mixers, receivers, upldown converters,
power dividers, VCOs, switches, filters and so on. Scalar network
analyser allows magnitudk measurements. It can be distinguishedfrom
a vector network analyser mainly in terms detection technique
employed. A diode detector is used in case of a scalar network analyser.
The diode detector has two operational regimes as shown in Fig. 8.13.
In the square law regime, output voltage is proportional to the input
power or square of input voltage. The response becomes linear above a
certain power level. The scalar analysers have the capability to extend
the usable dynamic range by providing compensation for change in
detector characteristics. For instance, SNA, Model No. 54000Afrom MIS
Anritsu (Fig. 8.14)has a dynamic range of greater than 70 dB (- 55 dBm
to + 16 dBm).
4
/'
"'
A

Linear
Region
Square Regi~n

Log. (Pin)
Fig. 8.13
There are two operational modes namely DC detection and AC
detection. In case of DC detection, excitation signal is the uninodulated
RF signal and the output is a DC proportional to input power. In case
of AC detection, excitation or test signal is a square wave amplitude
modulated R F signal (Fig. 8.15) and the output is a square wave. AC
detection has the advantage that it is not affected by detector's DC drift
and temperature sensitivities. A SNA usually employs a directional
bridge, as shown in Fig. 8.16, which combines in itself the detection and
350 MICROWAVES AND RADAR
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 35 1

signal separation functions, It works like a wheatstone bridge. With

reference to Fig. 8.16, If the test port resistance equals 50 R, voltage

DUT . II
n o
@I-
. .. . ... 4/ J
nUnUnL

Fig. 8.15

Fig. 8.16
null is measured. If it is not 50 Q, then a voltage proportional to
mismatch (deviation from 50 R) of DUT is measured. Directional bridge
has a broad band characteristics with a very good directivity perfor-
mance (typically 40 dB). But the 6 dB insertion loss present in the
incident RF path limits the power available at DUT and subsequently
the dynamic range of a transmission measurement.
Before making transmission measurements with the scalar net-
work analyser on a given device, it is important to establish the
magnitude reference of the measurement set-up. The commonly used
method is to use a simple Transmission Through response calibration
as shown in Fig. 8.17 by bypassing the DUT. This removes any signal
path loss differences in the reference and test paths. The Through
response establishes the 0 dB magnitude reference for the measure-
ment.
Incident
*
Receiver --, Display

DUT T *

Directional
Coupler ' ~hrough'
-
n
"
A

Fig. 8.17
352 MICROWAVES AND RADAR

Incase ofreflection measurements also, it is important to establish

the magnitude reference. In a simple reflection response calibration, a
short circuit as shown in Fig. 8.18 is normally used because the response
of a short (Reflection coefficient, T = 1L18Oo)is well known. Such a
calibration compensates for the differences in the measurement signal
paths. When the short is replaced by thr DUT, the return loss, equal to
(20 log p), is the difference between 0 dB reference line and the
measurement.

Receiver

Display

Short
Fig. 8.i8
8.6. Measuring With Vector Network Analyser
A vector network analyser (VNA)measures complex transmission
and reflection characteristics of microwave devlces as a function of
frequency. That is, it measures both magnitude as well as phase
information. It does so by comparing the incident signal with the
transmitted and reflected signals.
A vector network analyser can be distinguished from its scalar
counterpart primarily by the receiver architecture and complexity. The
basic block schematic arrangement is similar to the one shown in case
of a scalar analyser. The signal separation devices are splitters, couplers
and/or bridges. The receiver portion of a VNA is a multichannel receiver
capable of tracking the tunnel frequency. It employs either fundamental
or harmonic mixing. The broad band swept microwave signal is down
converted to a fixed IF that is independent of test frequency. The
instrument is capable of determining the magnitude of signal in each
receiver channel and phase relationship of any two receiver channels.
Due to its ability to measure phase characteristics, VNA is capable of
characterising complex impedances and phase delay characteristics of
devices under test. It is important that receivers are linear in their
conversion characteristics. The input signal should not be large enough
to cause any limiting or compression. Also, signals to receiver input
should not be so small that receiver sensitivity and accuracygets limited
by noise or cross talk. Therefore care should be taken to maintain
optimum signal level to DUT and receiver inputs.
The Transmission Through response calibration used for estab-
lishing the magnitude reference also establishes the phase reference in
case of transmission measurement with a vector analyser. Similarly,
measurement calibration with a short eircuit also establishes the phase
reference for reflection measurement. Reflection data can also be dis-
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 353

played in an impedance format. Reflection coefficient is related to '20'

. 'P and '20'are known, 'ZL' can be displayed.
and ' 2 ~ ' If
Other Network Analyser Measurements
In addition to the normal transmission and reflection measure-
ments that are carried out with network analysers, other important
measurements made with network analysers include Electrical Delay,
Group Delay and Distance-to-Fault.
Electrical Delay: All modern network analysers have an in-built
electricallength adjustment capability that can be used to vary the path
length of the reference channel. This adjustable phase delay introduced
by the device under test can be used to compensate for the phase delay
introduced by the DUT. Once the phase is balanced, the path length
difference added to achieve that is the electricallength of the DUT. Both
delay time as well as electrical length can be displayed.
Group Delay :Group delay is another important parameter that
can be measured with a network analyser. Group delay is the rate of
change of phase response as a function of frequency.
That is, Group delay = - d$/do
where o = Radian frequency
I$ = Phase in radians

Also, Group delay = - (1,360) d$/df where 'f is frequency in

Hertz.
A linear phase versus frequency relationship (Fig.8.19~) leads to
a constant group delay, the absolute value of which is then the transit
time through the DUT. Any deviation from this constant group delay
corresponds to deviation from linear phase (Fig. 8.19 b). Deviation from
constant group delay indicates distortion. Group delay is a measure of
transit time through the DUT for a given frequency. It is given by the
slope of the phase versus frequency curve at the frequency.

Fig. 8.19
The technique used for measuring group delay in modern network
analysers is to measure phase at two closely spaced frequencies and
then computing the slope as shown in Fig. 8.20. The two sample
354 MICROWAVES AND RADAR

frequencies however should be so chosen that the phase difference does

not exceed 180". Maximum measurable group delay range is then a
function of the frequency stability of the source. For instance, if the
source has a frequency resolution of 1Hz, maximum group delay range
would be 500 ms.

Fig. 8.20
Distance-to-Fault : The ability of network analyser to charac-
terise linear networks as a function of frequency can be put to use in
determining distance to fault in transmission structures such as cables,
transmission lines. By applying Fourier transform analysis techniques
to the frequency domain data, time domain information is obtained
which for a known propagation velocity in the transmission medium
can be transformed into distance information. The resultant output is
a useful tool for locating faults in transmission structures.
8.7. Microwave Power Measurement
Microwave power measurement is one of the most. commonly
carried out measurements. Power measurements at microwave fke-
quencies are usually divided into low (less 10 mW), medium (10 mW
to 1W) and high (greater than 1W) power levels.
The commonly used power sending elements or the detectors in
case of low power level measurement are crystal diodes, bolometers,
thermistors and thermocouples. In case of bolometers and thermistors,
the absorbed microwave power gets converted into heat energy which
in turn causes a proportional change in the resistance of the sensing
element. Thermocouples produce a DC voltage proportional to the
incident power.
The measurement of medium power levels is usually done by using
calibrated attenuators together with low level power meters (Fig. 8.21).
Thermocouples however, are being used at present for measurements
of power levels of about 100 mW without the use of any attenuators.

Power ;n 4 Calibrated
Attenuator Sensor
(
Fig. 8.21
MICROWAVE MEASUREMENTS AND INSTRUMENTATION -355

High power levels are measured by using directional couplers

orland attenuators together with low level power metres (Fig. 8.22).
Microwave calorimeter is another measuring instrument used for mea-
surement of higher microwave powers. In a microwave calorimeter, the
power level is indicated by the rise in temperature as the microwave
energy is dissipated in a lossy termination.

Sensor . * Meter

4
Microwave 1 X I ~ ~ Power
g h
Power in 1
L -----------
I Load
J
Direct~onal
Coupler
Fig. 8.22
There are two broad categories of microwave power detectors
namely the Thermal detectors and Diode detectors. Thermistors,
Bolometers,Thermocouples belong to the category of thermal detectors.
Thermal detectors are inherently slow. These detectors exhibit linear
characteristics over a dynamic range of - 30 to + 20 dBm. The dynamic
range of these detectors can be shifted upwards by adding an attenuator
a t the detector input. This also improves the measurement uncertainty.
Power meters incorporating these detectors provide considerable noise
filtering and averaging to compensate for the low sensitivity of the
voltage response. For very low power levels in the range of -30 to -10
dBm, the system tends to be very noisy. The power meter must be zeroed
by switching off the source of microwave power before performing the
measurement to eliminate effect of noise at low power levels. Thermal
based power meters come with a calibration source to which the ab-
solute response of the detector power meter combination can be ad-
justed.
Diode detectors are fast and have arelatively large dynamic range,
typically -70 to +20 dBm. Hot carrier, Crystal and Schottky barrier
diodes are the commonly used types. These detectors have two distinct
operating regions namely the Square Law Region where the output
voltage is proportional to square of input signal voltage level or linearly
proportional to input power level and the Linear Region where the
output voltage tends to go non-linear as a function of input power (Fig.
8.23).
The characteristics are typically linear in the range of -70 dBm
to -30 dBm as can be seen from the graph shown in Fig. 8.23.
Fig. 8.24shows relative comparisonof different types of microwave
power sensing measurement devices. Microwave power meters using
366 MICROWAVES AND RADAR

Fig. 8.23
diode detectors also come with a calibrator which can account f o ~
non-linearityin the detector's response for power levels above -30 dBm.
(Thermal) (Diode) (Mixing type
r

I
Fig. 8.24
Measurement by Comparison/Substitution
The basic underlying principle in a thermistor or bolometer based
microwave power measurement set-up relies on accuratelydetermining
a DC or a low frequency AC power using a method of comparison or
substitution with the unknown microwavepower. In a simpleconfigura-
tion, as shown in Fig. 8.25, the power sensing element is placed in one
of the arms of a Wheatstone bridge and exposed to DC or low frequency
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 357

AC bias power so as to balance the bridge. The same sensing element

is then exposed to the unknown microwave power which causes bridge
unbalance. The balance is restored by varying the bias power. This
amount of bias power applied to restore the bridge balance is then a
measure of the unknown power. The bridge balance is restored either
manually or by using a self-balance circuitry comprising of a high gain
differential amplifier in a feedback loop as shown in Fig. 8.25.

1 Meter

Fig. 8.25
Source of Error or Uncertainty
The main sources of error in a microwave power measuring set-up
including the following :
1. RF losses in the power sensor mounts
2. DC-to-microwave substitution errors
3. Errors due to mismatch
4. Instrumentation errors
5. Source instability
The errors due to the first two sources listed above i.e. RF losses
and DC-to-microwave substitution errors, are expressed as a figure-of-
merit called Effective Efjiciency (q)of the mount which is defined as the
ratio of the DC power dissipated in the element to the microwave power
dissipated within the sensor mount.
This is, q = (pdt/prf)
This does not include the effect of mismatch, if any, from source to
sensor.
Mismatch effect is included in another parameter called Calibra-
tion Factor which is defined as the ratio of substituted DC power in the
sensor mount to the microwave power incident upon the mount from a
matched '20' source.
That is, Calibration Factor, K is given by :
Substituted DC Power in the sensor mount
K=
Microwave power incident upon mount from a matched source
Calibration factor is the most widely used calibration value in
power measurement systems as i t takes into account all three major
358 MICROWAVES AND RADAR

sources of error. We can determine the calibration factor for different

types of sensor mounts in use and also for Directional coupler/Sensor
mount power standards. For a Directional coupler~Thermistormount
power standard, it can be determined for a terminated type or for a feed
through type. A terminated type (Fig. 8.26), where the thermistor
mount is connected to the side arm of a directional coupler, the calibra-
tion factor is defined as the ratio of DC power substituted in the
thermistor element of the mount connected to the side arm of the coupler
to the microwave power incident upon input to directional coupler from
a non-reflecting source with the coupler terminated in a matched load.
l herrn~stor Precision
Mount
-

Fig. 8.26.
In the feed through type arrangement as shown in Fig. 8.27, it is
the ratio of the DC power substituted in the thermistor element of the
mount attached to side arm of directional coupler to the microwave
power incident upon a non-reflecting termination connected to the main
arm output of the directional coupler. The unknown microwave power
is the ratio of the substituted DC power to the calibration factor of the
measurement set-up. Fig. 8.28 shows the photograph of a power meter
from W s Anritsu Wiltron that can measure microwave power in the
frequency range of 100 kHz to 90 GHz and is capable of measuring both
absolute as well as relative power levels. It is compatible to a variety of
power measuring sensors.

I Thermistor
Mount
Precision
DC Bridge I
! x --
--Thermistor
Standard

>-
Mount
Directional
- - ~ -

Coupler
Power

Fig. 8.27
8.8. Noise Measurement
Noise Factor, Noise Figure (which is Noise Factor expressed in a
different way) and Equivalent Noise Temperature are the different
 GP-IB
' ILOCAL
-.
WATT
-..MOM-
d b
-*--*a

dBfRUt

Fig. 8.28
360 MICROWAVES AND RADAR

parameters used to express the noise performance of devices and

circuits.
Noise Factor % is the ratio of the input signal to noise power to
the output signal to noise power. The noise factor when expressed in
decibels is the Noise Figure.
F = (SzlNi)/ (So /No) and NF = 10 log F
Tbc other important noise parameter is the Equivalent Noise
TemAwrat~,-e, Te. Temperature can be very convenientlyused to express
noisa powcr. The concept of equivalent noise temperature can be ex-
pla4nec?by saying that if the device under test, say an amplifier, is
terlrindmi at its input by a resistance at zero degree kelvin (so as to
be 11o.s ,less), then the noise power present at the output of the amplifier
wouid b the same as the noise power that would be present if the noise
frre eq~ivalentof that ampifier was terminated at its input by a
remtancc: at a temperature of 'Te'. The expressions that relate the
equivplmt noise temperature to the noise factors and the noise figure
defined above are :
Te = Tl (F- 1)where 'Fis the noise factor and 'Ti' is the absolute
temperature of the input termination
NF = 10 log [(Te/Tz) + 11
An ideal amplifier bas a noise factor of '1' and a noise figure of
0 dB.
Noise Measurement
One of the commonly used methods of measuring broadband noise
is the Y-factor technique. The technique relies on the use of two noise
aources having widely different but -knownnoise temperatures. Y-factor
is the ratio of the output power reading from the device under test when
the DUT input is exposed to a hot source (which could be room tempera-
ture or an active noise source) and the output power reading when the
DUT input is exposed to a cold source (typically liquid nitrogen).

or Y = [K(Te + Th). BI/[K. (Te + Tc).Bl= (Te + Th)/(Te + Tc)

where Te = Noise temperature to be measured
Th = Noise temperature of the hot source
Tc = Noise temperature of the cold source
K = Boltzmann's constant
A rearrangement of the above expression gives :
Te = (Th - Y . Tc)/(Y - 1)
The noise figure of the same device can then be expressed as :
NF = 10 log [(Te/To) + 11
Best results are obtained with the Y-factor technique when
'Te', 'Th' and 'Tc' are interrelated by :
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 361

Fig. 8.29 shows the basic test set-up used for measuring the noise
temperature of a receiver using Y-factor technique. The disadvantage
of Y-facor being non-linearly related to the noise temperature, par-
ticularly at millimeter and sub-millimeter wavelengths, can be over-
come by using a more convenient parameter called the R-factor
determined by using a chopper circuitry to provide changes from hot
loads to cold loads at a rate of few Hertz and defined as :
R = [{ (Th + Tc)/2)+ Te I/(Th - Tc)
The numerator here is the average noise temperature and
denominator is the difference between the noise temperatures of hot
and cold loads (Fig. 8.30).The above expression can be re-written as :
Te =RAT - (Th + Tc)/2

Deviceunder Noise Fig.

Test Meter

Fig. 8.29

Fig. 8.30
Since hot and cold temperatures are fixed and known for a meas-
urement, 'R' and 'Te'are linearly related. Assuming quadratic detection
of powers (noise temperature), 'R'is the ratio of the AC and the DC
output components of the dotected chopped signal. This permits a
linear, direct reading noise temperature meter. Fig. 8.31 shows the
photograph of one such Noise Temperature Meter from MIS Farran
Technology Limited particularly suitable for optimising radiometer
performance. The instrument operates when the radiometer input is
chopped between sources at two different noise temperatures. The only
signal required by the instrument is the detected radiometer output
with the conseauence that this instrument can be used to o~timise
receivers working at any frequency from RF to sub-millimeter. No
362 MICROWAVES AND W A R

external synchronisation between the chopper and the instrument is

necessary.
Solidstate noise diode is the noise standard or reference other than
the hot and cold noise standards mentioned above. When a diode is
reverse biased beyond the breakdown voltage, the noise power available
is fairly constant as a function of frequency. The excess noise ratio
(ENR) defined in case of such a noise source is expressed as :
ENR = 10 log [(Tb/To) - 11
where Tb = Equivalent noise temperature of the diode in
the ON condition
To = Room temperature
The impedance offered by the diode varies considerably as it is
switched from OFF to ON condition. In order to minimise the effect of
that, the output port of the diode is padded with an attenuator. Com-
mercially available diodes for the purpose have ENR in the 5-15 dB
range.

Fig. 8.31
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 363

Measurement Errors
There are two possible types of errors that can creep into the
measurement of noise temperature or noise figure. One source of error
originates from the test set-up itselfwhere as the other pertains to the
noise standards and the interface between the noise standard and the
device under test.
The errors pertaining to the first category can be eliminated or at
least minimised by taking the following measures :
1. The system should be operated 10 to 15 dB below its 1-dB
compression point in order to ensure that the test set-up does
not saturate by noise peaks.
2. Image and harmonic responses should be rejected by using
image reject mixers and suitable filters.
3. In case of use of hotJcold noise standards, switching time
between hold and cold references should be minimised.
4. Errors due t o temperature variations should either be
eliminated or corrected for.
The uncertainties arising out of noise standards and noise stand-
ard - DUT interface could be caused by the following factors :
1. Uncertainties about the absolute available noise power from
the standard :Noise standards are calibrated by the manufac-
turer under certain given test conditions. They also provide
calibration/correction data for the standards made by them.
This data must be consulted while using a given noise stand-
ard.
2 . Uncertainties or errors cause by variation in the source im-
pedance :Variation in impedance when the noise standard is
switchedfrom one state to the other, for instance, could change
the estimated available noise power from the noise source. It
could also change the noise temperature of the DUT such as
pn amplifier. It could also change the gain of the amplifier if
the DUT was the one.
3. The third source of uncertainty is caused by the adapters used
to accommodate different types of connectors. These can cause
additional VSWR uncertainties and additional loss between
the noise source and the device under test.
8.9. Phase Noise
Phase noise or Phase Jitter is animportant performance parameter
in many R F and microwave systems including Doppler radars,
Microwave communication links, Mobile radio and so on. In these
systems, there is often a situation where a weak signal, which is the
desired signal, needs to be processed in the presence of a strong
interfering signal which could originate from the phase noise side bands
of the strong interfering signal in the adjacent channel. In a typical
situation, the down converted weak signal in the IF band gets masked
by the down converted phase noise side band of a strong interfering
signal in the adjacent channel. In case of a doppler radar, the strong
364 MICROWAVES AND RADAR

interfering signal could be produced by reflections from strong station-

ary objects. Phase noise often defines the performance of these systems
which makes it necessary to measure and specify the same.
Phase Noise in Time Domain
Phase noise can be expressed in the time domain in the form of
time jitter. It can be expressed by :
$(t)= At(t) . 2 d T
where T = Time period of the signal
At ( t )= Time jitter
An oscilloscope can be used to determine, the phase jitter in the
time dqmain. The oscilloscope is ~xternallytriggered with the signal
(V cos a t ) .The same signal is applied to the DUT whose output as given
by V cos f&t + $ (t)lis displayed on the oscilloswpe screen. The instan-
L J
taneous time jitter At (t) is proportional to the phase jitter. Figs. 8.32(a)
and (b) show the test set-up and the relevant waveforms respectively.
This, of course, measures the residual phase noise as it detects only the
phase jitter added by'the DUT. If the oscilloswpe triggered by reference
signal, then the display would give instantaneous phase differerice of
the signal with respect to the reference signal.

Oscilloscope
DeviceUnder
Test

Fig.8.32
Oscilloswpe measurement of the type described above however
has a poor sensitivity. The test set-up of Fig. 8.33 has a higher sen-
sitivity. The phase detector produces a voltage output proportional to
phase difference between two RF inputs. The figure is self explanatory.
Another very sensitive technique of measuring phase noise in time
domain is the use of Time Interval Counter. The counter measures
average frequency of the signal under test over a given sample time
'3.Deviation of this frequency (= hf) from the centre value is plotted as
a function of sample time. A typical plot is shown in Fig. 8.34..
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 365

DUT

Reference
t
Phase
1
Det tctor

t
I

,Filter -= Oscillo~cOpt
&

-
Source

Fig. 8.33

t t

1 sec. 1 min. 1 hr.

Sample Time (TI -
Fig. 8.34.
Phase Noise in Frequency Domain
Frequency domain measurements of phase noise means the
spectral density distribution of phase fluctuation. It is expressed by:
~ $ 6 =7 &s u>
where Sq(n = Spectral density distribution of phase fluctuation
h s = RMS phase deviation per at Fourier frequency f
f = Fourier frequency
The test set-up is similar to the one shown in Fig. 8.33 except that
phase detector output feeds a base band spectrum analyser. The
spectrum analyser displays the spectral density distribution of voltage
fluctuation which can be translated into spectral density distribution
of phase fluctuation with a calibration factor.
Use of Frequency Discriminator as a phase detector is also very
common. The output of a frequency discriminator is a signal proportion-
al to frequency fluctuations and same when applied to a base band
spectrum analyser displays the spectral density distribution. The signal
source output may be applied directly to the frequency discriminator as
shown in Fig. 8.35(a) or it may first be down converted to an IF and
then applied to an IF frequency discriminator as shown in Fig. 8.35 (b).
A delay line-mixer combination can also be used in place of a frequency
discriminator. It makes a very simple broadband frequency dis-
criminator.
 Source Under
Frequency
Discriminator Hu Baseband
Analyser

-1
Test

Mixer IFFrequency
Discriminator
Source Under
Test
1 Reference
Source

Fig. 8.35
8.10. Time Domain Reflectometry
Time domain reflectometry is basically r, technique used for locat-
ing and analysing mismatches in high frequency transmission systems
such as cables, antennae, attenuators etc. This technique is based on
feeding a fast rise time voltage step to the component under test and
then analysing the reflections produced by any impedance mismatches
or discontinuities in the system under test on the screen of a sampling
oscilloscope. The analysis of reflections not only makes it possible to
identifv the location of each individual discontinuity but also the nature
of the discontinuity.

Generator

Fig. 8.36
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 367
One such test set-up is shown in Fig. 8.36. The pulse generator
launches a fast rise time voltage step onto the system under test. The
incident step and the returning reflections are observed on the oscillo-
scope. Since the voltage step propagates the system at a finite speed,
anv discontinuities or mismatches that are separated in distance
prGduce reflections that are separated in time. ~ i c discontinuity
h can
be identified separately on the oscilloscope display and its exact location
determined by evaluation of the corresponding time separation. Also,
the magnitude of impedance mismatch and its resistive, capacitive or

Driving Reflected Type of discontinuity

Waveform Waveform in transmission h e
-------
_h --
-- -
zo ---
+
---.

T ... z0 ----
J
7
RL>L
RLCZO
x---
--- ---
Fig. 8.37

inductive nature can be determined from the magnitude and shape of

reflection. The reflections obtained in the case of resistive, inductive
and capacitive mismatches are shown in Fig. 8.37. In case of resistive
mismatch,Z is the characteristic impedance of the system following the
discontinuity.
8.11. Spectrum Analysers
While an oscilloscope gives a time domain representation of the
input signal, a spectrum analyser gives a display of the frequency
spectrum of the input signal. While in an oscilloscope, X and Y axes
respectively represent time and signal amplitude, in a spectrum
analyser, these axes represent frequency and signal amplitude. In other
words, a spectrum analyser gives a plot of the amplitude of the Fourier
transform of the input signal. Fig. 8.38 shows both time domain repre-
sentation as seen on an oscilloscope as well as frequency domain
representation as it would be seen on a spectrum analyser in case of
some of the better known communication signals.
Spectrum analysers are particularly useful for examination of
signals related to RF and microwave communication equipment and
systems. It could be analysis of amplitude us frequency plot of a given
signal source, antenna or signal distribution system or even as a
diagnostic tool to establish compliance to the many emerging regula-
tions regarding both desired and undesired emissions or assisting the
digital designer in improving the RFIEMI performance of high speed
digital systems. This diagnostic role has assumed particular sig-
 368 MICROWAVES AND RADAR

nificance with proliferation of high speed digital computers in the

contemporary communication scenario.
11: the paragraphs to follow, we shall discuss the operational
principle, different types and importmt application oriented specifica-
tions of spectrum analysers.

Frequency
Time Domain
Domam

Slngle Frequency Signal

1 f

JllLL f

Fig. 8.38
-
Spectrum halyser Basic Principle
The most commonly used architecture in spectrum analysers is
depicted in Fig. 8.39. It is basically a superheterodyne receiver whose

Mixer -
Input lnput IF Log --c Detector -c Video
Attenualor -C Filler Section Amp Filter
I

I sweep4-1
Generator Display1

Fig. 8.39
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 369

local oscillator (first local oscillator in case it is a double conversion

receiver) frequency has been swept. Though the architecture is similar,
a spectrum analyser has a much larger frequency range, a swept local
oscillator and a different IF stage architecture.
The purpose of the input attenuator is to limit the input signal
power so as to keep it within the normal operating range of the
instrument. Most of the spectrum analysers can handle 0 to 10 dBm at
their mixer inputs. The input attenuator keeps the larger signals within
the acceptable range and also optimises the measurement dynamic
range of the instrument. Also, under no conditions, should the input
power level to the instrument exceed the power handling capability of
the input attenuator which is typically 0.5 to 1watt.
The input filter provides image frequency rejection. It is important
to avoid the undesired mixing products which may occur due to signals
a t image frequency. It is a low pass filter in a typical RF spectrum
analyser. In case of microwave spectrum analysers, tracking filters are
used. In the spectrum aaalysers having frequency range of above 50
GHz, where physical realisation of filter is difficult, unpreselected
mixers are used and the instruments offer facility to identify the desired
signal.
The IF stage largely decides the resolution bandwidth of the
instrument i.e. how close can the two simals get to each other in
frequency and still be seen as separate resp&ses. ' h e primary function
of the IF stage is to offer a wide selection of resolution bandwidth filters.
The narrower the filter, higher the resolving capability. The filter
selectivity is usually measured in terms of shape factor which is nothing
but ratio of 60 dB and 3 dB bandwidths. Sometimes, i t is taken to be
the ratio of 60 dB and 6 dB bandwidths. Equally important is the filter
type. While superior shape factor allows for better resolution of closely
spaced signals, the filter type influences the time domain performance
and better time domain performance allows faster sweep speeds with
better amplitude accuracy.
The Log amplifier compresses the input signal amplitude range.
It has a dynamic range in excess of 100 dB. The detector is an envelope
detector of the type found in AM communicationreceivers. The dynami ,-
range requirement on the part of the detector is not high due to ?he
range compression that occurs in log amplifier. In some spectrum
analysers, large dynamic range linear detectors like synchronous detec-
tors are used. The detector in that case is followed by a DC logarithmic
amplification to get display with desired dynamic range. Video filter
provides post filtering or averaging of the detected output. The filter
bandwidth is usually set to be small fraction of the selected resolution
bandwidth unless the measurement needs averaging.
The swept local oscillator is a very important component of the
spectrum analyser. The stability and spectral purity of the local oscil-
lator are important. A good local oscillator should have as little a
residual FM as possible. Infact, the oscillator stability sets the lower
limit for the resolution bandwidth achievable in a spectrum analyser.
370 MICROWAVES AND RADAR

The signal gets badly smeared if the resolution bandwidth goes finer
than this limit.
Modern spectrum analysers are microprocessor controlled. In the
older versions of the equipment, the parameters like frequency span,
sweep speed, resolution, which are so vital to maintaining the integrity
of measurement, needed to be adjusted manually by the user. Optimum
setting of these parameters is key to making correct measurement. In
the present day spectrum analysers, all this is taken care of by the
microprocessorinside the instrument. Most of the pperational functions
in these spectrum analysers are microprocessor controlled with the
result these functions are switch selected rather than vernier adjusted.
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 371

Resolution as mentioned in the earlier paragraphs is the ability of

the spectrum analyser to display discrete frequency component within
a frequency span. The capability to do so is a function of the spectrum
analyser bandwidth, sweep speed, frequency span and incidental FM.
Frequency span and sweep speed are usually selected to provide mini-
mum resolution bandwidth setting for a given CW signal as minimum
resolution bandwidth enhances sensitivity or SIN ratio. Also as the
sweep speed is increased, a critical rate is reached where both sen-
sitivity and resolution are degraded. Therefore, sweep time for a
calibrated display is dependent on resolution bandwidth and frequency
span. These parameters are interrelated by :
Sweep time = Frequency span/(Resolution and width)'
Fig. 8.40 shows the photograph of a typical spectrum analyser. The
baisc instrument (Model No. MS710 from M/s Anritsu Wiltron) has
frequency range of 10 KHz to 23 GHz which can be extended upto 140
GHz with an external mixer.
Front Panel Controls
Important controlsfound on the front panel of majority of spectrum
analysers are briefly deskribed below.
Intensity, Focus and Beam Alignment : Intensity controls the
brightness of the CRT trace. Trace alignment and Beam focus are
internal adjustments and are usually not available on the front panel.
Base Line Clip : When this control is activated, the baseline of the
display is clipped or subdued to enhance contrast between the display
and the baseline. This control is particularly valuable while viewing
pulsed spectra. An attempt to brighten the spectra in this case also
brightens the base line.
Readout :It can be used to switch the CRT readout ON,or OFF.
All spectrum analyser parameters are read out except TimeIDiv which
selects sweep speeds. The brightness of the readout is proportional to
the trace brightness.
Graticule Illumination :It switches graticule light ON or OFF.
Trigger :The trigger modes that are usually available with almost
all spectrum analysers include INT, FREE RUN, LINE and EXT. In
FREE RUN mode, the sweep is free running. When selected, all other
triggering modes are cancelled. In this mode, the sweep will not
synchronise with any input signal. In INT trigger mode, sweep is
activated internally from the vertical or video signal. In LINE, it is
triggered a t the line frequency rate of the power supply when a sample
of the AC power line voltage is used to trigger the sweep. In EXT, the
sweep is activated by an external signal applied to EXT IN input.
Triggering is usually FREE RUN for spectrum displays. However it
may sometimes be desirable or even necessary to trigger the display
when the event is time related to some source or when the frequency
span has been reduced to zero for time domain analysis.
372 MICROWAVES AND RADAR

Single Sweep : In the single sweep D ode, ona sweep is initiated

after the sweep circuit i; triggered.
Manual Scan : This control can 3e used to manually scan the
spectrum by keeping the TimeIDiv selector in the MANUAL position.
Manual scan is used to examine a particulhr point or pnrtio~lof the
display such as one of the null points of a frequency modulation
spectrum or where a slow sweep of a full scan would take unnecessarily
long time.
TimelDiv : It selects sweep rates with the slowest and fastest
sweep speeds depending upon instrument type. This control also selects
AUTO, MANUAL and EXT sweeps. In AUTO mode, the sweep rate is
selected by the internal microprocessor to maintain calibrated display
for any setting of Freq SpadDiv, Resolution bandwidth and Video filter
combination. In MANUAL position, as mentioned earlier, the horizon-
tal axis can be swept with the manual scan. With TimeDiv control on
EXT, an external ramp can be used to scan horizontally.
Frequency :This control tunes the center frequency.
CAL :When activated, the frequency readout can be calibrated to
center frequencyby adjusting the FREQ control for the correct reading.
It is deactivated after the calibration is over.
Degauss : It is used to minimise hysteresis effects and enhance
center frequency and display amplitude accuracy.
Auto Resolution : This control activates automatic bandwidth
selection for the chosen FREQ Span/Div, TimeDiv and Video filter
settings. An internal microprocessor selects bandwidth to maintain a
calibrated display. When TimeDiv is in AUTO, resolution bandwidth
becomes a function of Freq SpadDiv selection.
Zero Scan : With zero scan activated, the display reads out
Time/Div instead of Freq Span1 Div. The instrument can then be used
to display the time domain charateristics of the signal within the
capabilities of the selected resolution bandwidth.
Frequency SpanlDiv : The control selects Frequency span/Div.
Other than the zero scan, the minimum and maximum spanldiv depend
upon instrument bandwidth.
Vertical Display :This control selects the vertical display mode in
terms of dBldiv along the vertical axis. There are some other selections
also. A typical selectionis LINin which alinear vertical display between
zero volt (Bottom graticule line) and the reference level (Top graticule
line) is selected.
Digital Storage : It is used for storage of display information if
desired. Spectrum analysers have various storage modes enabling the
user to view desired feature of the spectrum.
Min RFAtten : Once activated, the RF attenuation is set to the
a i m u m . If deactivated, RF attenuation is set by the internal
microprocessor according to the algorithm selected by Min noisemin
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 373

distortion. The microprocessor assumes that the MIN RF ATTEN

selection is the minimum attenuation required for the expected signal
levels. It does not reduce RF attenuation below this level.
Reference Level :A change in REF LEVEL control changes the
display reference level which is the amplitude represented by top CRT
graticule.
Video Filter :The video filter restricts the video bandwidth SO that
noise or beat signals are reduced. When signals are closely spaced, the
filter may be useful in reducing modulation between two signals so that
they can be more easily analysed. The filter can also be used to average
the envelope of the pulsed RF spectra that has a relatively high PRF.
FFT Spectrum Analysers
FFT spectrum analysers like the one described above take a time
varying input signal and display its frequency spectrum. But it works
in an entirely different manner. It is not based on sweeping the local
oscillator of a superheterodyne receiver. Instead, it computes the fie-
quency spectrum by taking Fourier transform of the time domain signal.
According to Fourier theorem, any waveform in time domain can be
represented by weighted sum of sines and cosines. The FFT spectrum
analyser samples the input signal, computes the magnitude of its sine
and cosine components and displays the spectrum of these measured
frequency components.

Fast Four~er
D~splay
Transform (FFT)

Fig. 8.41
Fig. 8.41 shows the basic architecture of an FFT spectrum
analyser. The input signal is sampled at a high sampling rate similar
to what we do in a digitizing oscilloscope with due consideration to the
Nyquist sampling criterion according to which the samplingrate should
at least be twice the highest frequency component in the input signal.
The resulting digital record is then mathematically transformed into a
frequency spectrum using an algorithm called Fast Fourier Transform
(FFT). The resulting spectrum shows the frequency spectrum of the
input signal.
Fourier analysers offer fast and high resolution spectrum analysis.
These are much faster than their traditional analog counterparts by
one to two orders of magnitude due to the fact that they measure all
frequency components simultaneously and not one at a time. They are
particularly useful for spectrum analysis of low frequency signals (less
than 100 kHz) or where very fast measurements are desired. Signal
frequency components as closely spaced as several microhertz can be
resolved and accurately measured. Simultaneous measurement of mag-
nitude and phase on two or more channels provides high quality
network measurements.
374 MICROWAVES AND RADAR
MICROWAVE MEASUREMENTSAND INSTRUMENTATION 375
Fig. 8.42 shows the photograph of one such instrument from M/s
Stanford Research Systems. It is Model SR760 having a DC to 100 kHz
bandwidth and a dynamic range of 90 dB.
8.12. Frequency Counter
The most basic f u n k o n of a frequency counter is the measurement
of an unknown frequency. Modern frequency counters however offer
much more than just the frequency measurement. The other related
parameters such as Time Period, which is reciprocal of frequency, Time
Interval between two events and Totalise Count which is nothing but
curnulati-~ecount over a known period are the other functions that are
available with the present day frequency counters. These instruments
offering a variety of measurement options particularly two channel
functions such as time interval measurement are usually referred to as
Universal Counters.
Universal Counters - Functional Modes
The functions available with modern universal counters other
than measurement of an unknown frequency are (i) Time interval
measurement ( i i ) Period (iii) Time interval average (iv) Totalise ( v )
Frequency ratio A / B (vz) Phase A rel. to B (vii)Pulse width etc.
Time Interval Measurement :This mode measures the time that
elapses between the occurrence of two events. One of the events called
the start signal is usually fed into one of the channels while the other
called the stop signal feeds the second channel. Resolution of measure-
ment is typically 100 ns or better. A typical application of this measure-
ment mode is in determination of propagation delay in logic circuits.
Variations of this mode can be used to measure pulse width, rise and
fall times.
Time Interval Average :This mode can be used to improve meas-
urement resolution in the time interval measurement mode for a given
clock frequency. The resolution improves as square root of the number
of measurements. That is, average of 100 measurements would give a
10-fold improvement in resolution.
Period : In this mode, the time period of the input signal is
measured by counting counter clock between two successive leading or
trailing edges of the input signal. Again period average function can be
used to improve upon measurement resolution for a given clock. For
instance, if the measurement is done for 100 periods instead of one
period for a given clock frequency, the measurement resolution would
also improve by a factor of 100.
Totalise :Totalise mode gives a cumulative count of eventa over a
known time period.
Frequency Ratio A I B : This gives the ratio of the frequencies of
signals fed to A and B channels. This feature can be used to test the
performance of prescalers and frequency multipliers.
Phase A re1 to B : It compares the phase delay between the two
signals with similar frequencies.
376 MICROWAVES AND RADAR

Basic Counter Architecture

Fig. 8.43 shows the architecture of a frequency counter when it is
being used in the frequency measurement mode. The oscillator section
comprising of a crystal based oscillator and a frequency divider chain
generates the clock pulses. The clcck pulses are used to trigger a flip
flop vrhose output serves to enable or disable the AND gate. When the
AND gate is enabled, the input signal after passing through the signal
conditicning sectiou comprising of level shifting amplifiers, com-
parabrs ctc. reaches the counter. In the simplest case, if the AND gate
is e.lshl~xlfor 1s (which is the case when the flip flop clock input is 1
Hz).tl.en the counter count will represent the signal frequency. The
mee-;urementresolution in this case would be 1Hz. The measurement
resoiu?ion can be improved by enabling the AND gate for a longer time.
For in3tance, a 0.1 Hz clock at the flip flop input would give a 10 s gate
tiil~eand a consequent 0.1 Hz resolution. Similarly a shorter measure-
ment for a gate time of 0.1 s (corresponding to clock of 10 Hz)gives a
rnexrlement resolution of 10 Hz.

T m e base
Usc~llator Flip-Flop Display
.I

I I
Condit~oner
11!11111111111 Counter
1
Fig. 8.43

AND
gate
Time base 31videi
Oscillator

Fig. 8.44
The same building blocks when slightly rearranged
Fig. 8.44 can be used to measure time period.
the AND gate is now determined by the frequency of
and not by the clock frequency. The number stored in
is proportional to the number of clock pulses
during the period of the input signal. The same
time interval (TI) measurement by having two
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 377

with one enabling the AND gate by say setting the flip flop and the other
disabling the same by resetting the flip flop.
Reciprocal Counters
The rechprocal counter overcomes some of the limitations of the
basic counter described above, the most important of all being inade-
quate resolution particularly when measuring low frequencies. The
basic counter had a resolution of 1 Hz for a gate time of 1s and resolution
could be enhanced only by increasing the gate time. If the gate eime
could not be increased in a certain application, the resolution gets
restricted to 1 Hz. The basic counter measures frequency in terms of
event count. Depending upon the gate time which 1s 1s or any other
decade value such as 10,100 etc. the decimal point appropriately placed
in the count gives frequency. What is important to note here is that
computation of frequency involves computation of event count only. The
frequency which is given by event count divided by time taken is
calculable from event count itself if the time is 1s, 10 s, 100 s etc.
In the reciprocal counter, both events as wel: as time are computed
and the ratio of the two gives the frequency. Advent ofreciprocal counter
was made possible due to availability of digital logic that could perform
arithmetic division econormc:dly and with precision. Fig. 8.45 shows the
reciprocal connter hardware 7 he processor is the heart of the counter
hardware and controls almost every other building block. The
synchronising logic block routes the A and B channel inputs and the
time base signal to the Event and Time counters. The routing is
determined bv the measurement function. The com~utationsare done
in the proces~orblock.

Processor
I\f I Routing Logic I 1 -,
1 I

Fig. 8.45
As a matter of comparison, let us see as to how the two counters
having an internal clock of 10 MHz would respond to.measurement of
a signal frequency of 50.38752 Hz. The basic counter will display 50 Hz
assuming a gate time of 1s as the event count will be 50. The reciprocal
counter will also have an event count of 50 but it will also measure time
378 MICROWAVES AND RADAR
with a resolution of 100 ns (for 10 MHz clock) equal to 0.9923328 s. The
measured frequency will therefore be 50.38752 Hz. The frequency
resolution offered for 10 MHz clock is 7 digits equal to 0.000005 Hz in
the present case for a 1s gate. The resolution could be further enhanced
by increasing the clockfrequency.Since clock frequencies upto 500 MHz
are practical, a reciprocal counter would give a resolution of 2 ns for a
1s gate time.
The resolution of reciprocal counters can be further enhanced by
using a technique called interpolation. It is possible to achieve a 9-digit
resolution with a 10 MHz clock using interpolation techniques which
otherwise would require a clock of 1GHz. Higher resolution is impor-
tant, when we are looking for a given resolution in shorter gate times.
The details of interpolation techniques are beyond the scope of this text.
Continuous Count Counters
The counter architectures discussed in the preceding paragraphs
had a counter that counted for a known period equal to the gate time.
These counters have a dead time when the gate is disabled. Such
counters could miss vital information that codd be important to the
measurement. The continuous count counter architecture is based on
the fact that if different measurements of a certain parameter of a signal
are not disjoint and the relationship that they have is made use of, the
measurement resolution can be significantly enhanced by applying
what we call as curve fitting algorithms. These counters have all the
attributes of reciprocal counters with the additional ability of reading
the event, the time and the counter without having to disable the gate.
Counter Specifications
The data sheets and manuals of universal counters contain
detailed specifications of the instrument. The important ones include
Sensitivity, Bandwidth, Resolution, Accuracy and Throughput.
Sensitivity: It refers to the smallest signal that the instrument can
measure and is usually expressed as mV(rms)or p-t-p. Sensitivity of 10
mV(rms) to 20 mV(rms) is typical. In majority of measurement situa-
tions, sensitivity is not the issue.
Bandwidth : The bandwidth of the counter is its front end
bandwidth and is not necessarily the same as the maximum frequency
that the counter is capable of measuring. Measuring a signal frequency
higher than the instrument's bandwidth only reduces its sensitivity
specification and requires a larger minimum input signal. However, the
bandwidth does affect the measurement accuracy in case of some
parameters, rise time being one such parameter. Thus it is always
preferable to choose a counter with as high a bandwidth as possible.
Bandwidth is not explicitly mentioned in the specifications. However,
it can be guessed by looking at variation in sensitivity across the
frequency range of the instrument.
Resolution : Resolution refers to the minimum frequency incre-
ment in case of frequency measurement and time increment in case of
time interval measurement that can be resolved. The resolution is
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 379

usually very close to the least significant digit and is often +1 count of
LSD. Noise in the input signal, noise in the front end and input signal
slew rate are some of the factors that affect resolution.
Accuracy : Accuracy is related to resolution but not the same as
resoluticn. Factors like Time base (or clock) accuracy and Trigger
accuracy must be considered alongwith resolution specification to deter-
mine ultimate accuracy of frequency measurement. Time base error
affects measurement accuracy according to :
Frequency accuracy =Resolution f Time base error x Frequency
Trigger level accuracy is the precision with which trigger level can
be set. If there is an error in the trigger level setting, the trigger timing
is changed thus affecting measurement accuracy.
Throughput :Throughput is related to resolution. For instance,
increasing the gate time of a certain frequency measurement increases
the measurement resolution by the same factor but it slows down the
throughput by almost the same amount. Other factors affecting the
throughput are more related to the speed of the microprocessor and the
interface system. Two factors to watch here are the number of meas-
urements the counter can deliver through the interface and the speed
with which the counter can switch between different functions or
set-ups. If short gate times are being used andlor measurements are
being switched between different functions repeatedly, these factors
become important.
Microwave Counters
The counter architectures discussed in the preceding paragraphs
(conventional, reciprocal, continuous count) are usually good enough
upto 500 MHz or so. Counters meant for carrying out measurements at
frequencies at RF beyond 500 MHz and microwave frequencies employ
a different architecture. There are two types of architectures in use for
building nlicrowave counters. One uses a prescaler while the other is
based on down conversion.
Prescaler Counters use a prescaler placed between the front end
and the gating circuitry of the counter. Infact, prescalers are available
inside the counters as an optional channel to extend the frequency range

LPF - - IF
Ampl~f~er
Rec~prccal
Counter

I
Step
Recovery
Diode
- - Local
Oscillator
I

Microproce-
ssor Control
'D'spP'

Fig. 8.46
MICROWAVES AND RADAR
MICROWAVE MEASUREMENTS AND INSTRUMENTATION
of measurement. Extension upto 3 GHz is typically available with a
prescaler. Prescalers are not used with pulsed microwave counters due
to their tendency to self oscillate. When used with basic counter,
prescaler causes degradation of resolution. It is because frequency
resolution of basic counter is dependent upon contents of the event
counter and due to location of the prescaler before the gating circuitry,
its contents can not be read. The resolution is not affected when the
same is used in a reciprocal counter.
In the Down Converter based microwave counter, the input signal
frequency is down converted and an IF produced. IF, which is the
difference between the input signal frequency and the local oscillator
frequency is then counted. The actual frequency is then computed from
(LO + IF). Covering a frequency range of tens of GHz for an LO is an
expensive proposition. The solution is to use a relatively lower frequen-
cy LO (approx. 200 MHz). The LO drives a step recovery diode that
produces a sharp pulse with usable harmonics upto the desired range.
This pulse drives a sampler which samples points of the input signal.
The resulting IF is low pass filtered and counted. The actual input
frequency is then given by (N x LO +IF) where (N) tells as which
harmonic of LO goes through mixing operation. One of the methods to
determine (N)is to measure IFs at two slightly different LO frequencies.
(N) is then given by (IF1- IF2) 1(LOI - LO2). But all this is instrument's
headache and may take several tens of milliseconds only. Fig. 8.46
shows the architecture of a microwave counter.
Pulse microwave counters use the architecture described above
with the additional gating circuitry to position the gate comfortably
within the pulse. Fig. 8.47 shows the photograph of microwave frequen-
cy counter (Model MF76Afrom M/s h r i t s u Wiltron)having a frequency
range of 10 Hz to 18 GHz. In this instrument, measurement of signals
from 10 Hz to 200 MHz is direct while signals from 200 MHz to 18 GHz
are measured with the help of a highly stable internal heterodyne
converter. The instrument is microprocessor controlled. Measurements
on pulsed microwave signals in the frequency range of 100 kHz to 18
GHz are also possible with this instrument. It has a dynamic range of
-22 dBm to + 10 dBm.
8.13. Frequency Synthesisers and Synthesised
Functionlsignal Generators
Frequency synthesisers generate sinusoidal signals of extremely
high frequency stability and exceptional output level accuracy. Fre-
quency synthesisers and similar instruments such as synthesised func-
tionlsignal generators a r e used to provide t e s t signals for
characterisation ofdevices, subsystems and systems. Synthesisedfunc-
tion generators in addtion to providing spectrally pure and accurate
CW sinusoidal signals also prnvide other waveforms such as ramp,
triangle, square and pulse. Synthesised signal generators in addition
to providing spectraly pure and accurate CW signals also have modula-
tion capability and can be used to generate AM, FM, PM and pulse
382 MICROWAVES AND RADAR

modulated signals. There is another class of synthesised function gen-

erators called synthesised arbitrary waveform generators. Majority of
synthesised function generators have a limited arbitrary waveform
generation capability built into them. However, these are available as
individual instruments also.
All the above mentioned instruments have one thing in common
and that is synthesis of the signal which lends ultrahigh frequency
stability and amplitude accuracy to the generated waveform. They
therefore have more or less similar architecture for a given technique
used for frequency synthesis.
Direct Frequency Synthesis
Frequency synthesiser in its basic form uses a reference oscillator,
which is an ultrastable crystal oscillator, and other signal processing
circuits to multiply the oscillator frequency by a fraction hUN (where M
and N are integers) to generate the desired output frequency. One

L j j , z Fig. 8.48

such arrangement is shown in Fig. 8.48. It comprises of an assortment

of frequency multipliers and dividers,, mixers and band pass filters
(BPFs). The diawam shows the use of this architecture to generate 17
MHz. In this arrangement, if the BPF had a passband centered around
3 MHz, the output would have been 3 MHz as the mixer moduces both
sum and difference components. This method of frequeky synthesis
has several disadvaatages more prominent being the technique being
highly hardware intensive and therefore expensive, loss of phase con-
tinuity while switching frequencies with the result that this technique
has not found favour with the designers of frequency synthesisers.
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 383

Indirect Synthesis
In the indirect synthesis, the output is not directly derived from
the quartz crystal based reference oscillator. Instead, the reference
oscillator is used in a phase locked loop wired as a frequency multiplier
to generate an output frequency that is M/N times the reference
oscillator frequency. The output is taken from the VCO of the phase
locked loop. Fig. 8.49 shows the basic arrangement. If we insert a
divide-by-Ncircuit between the reference oscillator and phase detector's

Reference
Oscillator
D~vlde
by N
- Phase
Detector

4
'LPF

-
Fig. 8.49

VCO -
Ref Osc
(frei)
t
M~xer
0

fN l
r
1
Mult~pher Phase Loop
-) 'N2 -)
PLL ( x M ) Detector F~lter
A v
Summlng PLL
Modulus
(MI

Fig. 8.50
384 MICROWAVES AND RADAR

signal input and a divide-by-M circuit between the VCO output and
phase detector's VCO input, then the loop will lock with the VCO output
as fief,< (M/TJ). The frequency resolution of this architecture is x
(fref/N). 'frrf is the reference oscillltcr output frequency. The loop
frequency switching speed is of the order of ten times the period of
reference frequency input to loop phase detector. That is, if we desired
a frequency resolution of 1 Hz, the switching time is going to be of the
order of 10 s which is highly unacceptable. Another disadvantage of this
architecture is that frequency multiplier loops also multiply noise at
the phase detector which manifests itself in the form of noise sidebands
at VCO output. This --e.jtricts the maximum multiplication factor to
few thousaada in this srrangexent with the result that resolution gets
limited. In case finer resolution is needed, sequences of multiplication,
division and addition are used involving more than one phase locked
loops. One such arrangement is shown in Fig. 8.50. The synthesiser
output in this case is given by :
frefx MINI Nz}+ 1 1
This technique can be extended tc, get any desiredresolution. Since
the n~ultiplicationnumbers are low and the loop frequency is high, the
output will have low noise sidebands. Also, the synthesiser is capable
of fast frequency switching.

Modlllus t
Control e N or
-)
(N+1)

1
fret
= Phase
Detector - Analog
Adder

17
Converter

Fig. 8.5 1
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 385

Another popular method of indirect synthesis is Fractional N

synthesis where a single PLL is made to lock to the non-integer multiple
of loop reference. This technique can be used to achieve frequency
resolution of the order of microhertz at switching speeds of the order of
a millisecond or so. Fig. 8.51 shows the basic architecture. The con-
figuration functions as follows :
The integer part of the desired multiplier is supplied to the digital
divider placed between VCO output and phase detector in the form of
its dividing factor. The fractional part is supplied to the accumulator.
The accumulator is clocked by the reference source derived from crystal
oscillator. The quantum of fractional input is added to the accumulator
contents every clock cycle. The VCO output is N x F times the reference
input when the loop is locked. The circuit functions in such a way that
the contents of the accumulator predict the expected phase detector
output resulting from frequency difference of the two phase detector
input signals. The D/A convertor is then so scaled and polarised that
the D/A output waveform cancels the phase detector output waveform.
The two waveforms are added in the analog adder, sampled and filtered
to provide the oscillator control voltage. Also, to keep the phase detector
output within its linear range, whenever the phase difference between
the two inputs to the phase detector tends to become 360which is the
maximum the phase detector can tolerate without going out of range,
the phase of the divider output (which is ahead of reference input in
phase) is retarded by 360-y either changing the divider modulus to
(N+l)momentarily or by any other means. In the architecture shown,
the modulus is changed to accomplish this on receiving a command from
the BCD adder a t the time of accumulator overflow.
Sampled Sine Synthesis (Direct Digital Synthesis)
This method of frequency synthesis is based on generating the
waveform of desired frequency by first producing the samples as they

Frequency
n a n t 4 IC

Sine Look up
table,with inter-
polat~onbetween
stored entries
- DlA
Conve

Fig. 8.52
386 MICROWAVES AND RADAR
would look like if the desired waveform wa., sampled or digitised
according to the Nyqulat sampling theorem, and then inberpolating
amongst these samples to construct the waveiorm. -4s the frequency is
rate of change of phase, this information is made use of to generate
samples. Sine of different phase values are stored in a memory which
is addressed by phase increment information stored in an accumulator.
Fig. 8.52 shows the simplified block schematic representation of direct
digital synthesis. When the accumulator is clocked at a fixed frequency,
the contents of the accumulator jump by the phase increment whose
digital equivalent information is stored in phase increment register
(PIR). By changing the contents of the PIR, the output frequency can
be changed. The rate at which the look up table in the memory is
addressed is given by clock frequency and phase increment during one
clock period is given by PIR contents. For instance, if the contents of
FIR represented a phase angle of 36" then the digital samples present
at the output of the memory would correspond to phase differences of
36", 72 ", 10B0,144", 18O0,216", 252", 288", 324" and 360" to complete
one cycle of output waveform. The ten samples will be produced in ten
clock cycles. Therefore, the output frequency will be one-tenth of the
clock frequency. In general, output frequency is given by :
1 X &lock
where $ = phase increment in radians
The digital samples are converted into their analog counterparts
in DIA convertor and the analog samples interpolated to construct the
waveform. The interpolator here is a low pass filter. The relevant
waveforms are shown in Fig. 8.53.

Fig. 8.53
This method of synthesis derives its accuracy from the fact that
both the phase increment information and the time in which phase
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 387

increment occurs can be computed to a very high degree of accuracy.

With frequency being rate of change of phase, the resulting waveform
is highly stable. The most important feature of this technique however
is its capability to provide instantaneous switching. This is possible
because size of angle increments between two consecutive table look-
ups may be changed instantaneously. The limitations of this technique
include Quantisation noise and Aliasing inherent to any sampled data
system. Another serious disadvantage is that of presence of spurious
components due to imperfections and inaccuracies in D/A convertor.
The highest frequency that can be synthesised is limited by the maxi-
mum speed of the available digital logic. The usable frequency range of
direct digital synthesis output may be extended by a variety of techni-
ques. However, depending upon the technique used, some of the ad-
vantages of the DDS may be lost. Just as with more conventional
synthesisers, the DDS output may be doubled. mixed with other fixed
sources or used as a reference inside a PLL.
Specifications
Frequency range, Resolution, Frequency switching speed and Sig-
nal purity are the important synthesiser specifications.
While considering the Frequency Range, it is important to note
whether the claimed frequency range is being covered in a single band
or a series of contiguous bands. This aspect is significant from the
viewpoint of noise performance which may be different in different
bands in case the frequency range is covered in more than one bands.
This often leads to a larger transient when the frequency switching
involves switchover of the band also. Frequency resolution is usually
the same throughout the range. I t is typically 0.1 Hz though resolution
as fine as 1pHz is also available in some specificinstruments. Frequen-
cy synthesiser, Model No. 68000B from M/S Anritsu Wiltron (Fig. 8.54)
has a frequency range of 10 MHz to 67 GHz and a frequency resolution
of 0.1 Hz.
Frequency Switching Speed is a measure of the time required by
the source to stabilise at a new frequency after a change is initiated. Ir
the PLL based synthesisers, it depends upon loop's transient respon :,.
characteristics. Switching time is typically several hundreds of
microseconds to tens of milliseconds in PLL based synthesisers and a
few microseconds in instruments using direct digital synthesis (DDS)
technique.
Signal Purity tells how well the output signal approximates the
ideal single spectral line. Phase noise is one parameter that affects
signal purity. It refers to the sidebands which result from phase
modulation of carrier by noise. It is specified as the total sideband power
in dB with respect to carrier ( B c ) . Presence of spurious signals result-
ing from undesired coupling between different circuits within the in-
strument and distortion products in signal mixers also spoil signal
purity. For the instrument shown in Fig. 8.54, spurious levels are less
than - 60 dBc and the SSB phase noise is less han - 80 dBc / Hz.
388 MICROWAVES AND RADAR
390 MICROWAVES AND RADAR

Synthesised Function Generators

Synthesised function generators are function generators with the
frequency precision of a frequency synthesiser. The hardware of a
synthesised function generator is similar to that of a frequency syn-
thesiser with additional circuitry to produce pulse, ramp, triangle and
squara functions. These instruments with additional modulation
capability are referred to as Synthesised Signal Generators. Fig. 8.55
shcws the photogaph of a Synthesised Signal Generator (Model No. MG
3602 A1 from hI/s Anritsu Wiltron. It has a frequency range of 100 kHz
to 308C hiHz with SSB phase noise less than -123 dBcJHz and residual
Fivl lesr than 4 Hz (at fc = 500 MHz).
Direct digital synthesis described in the earlier pages of this
chapter i s almost invariably used in synthesised functionhignal gener-
atcr design. Advances in digital technology have made DDS based
synthesised functionlsignal generators truely versatile. Synthesised
sin? v ave output upto 30 MHz and other functions such as pulse, ramp,
trianglz etc. upto 100 kHz, all with a resolution of 1microhert~,are
ava!able i;l contemporary synthesised function generators.
MICROWAVE MEASUREMENTS AND INSTRUMENTATION 391

Review Questions
1. Briefly describe the important transmission and reflection
parameters as relevant to testing of various microwave
devices. How are they related to scattering parameters ?
2. What is a Network Analyser ? Differentiate between a scalar
and avector network analyser. Which parameters can usually
be measured effectively with a vector analyser ?
3. How do we measure microwave power by the method of sub-
stitution ? With the help of a typical set-up, describe how we
can measure microwave power.
4. Which power measurement set-up you would recommend
when the power under measurement falls in the range of
( a ) 1mW to 10 mW ( b ) 100 mW to 1W ? Justify your answer.
5. Where do we use time domain reflectometry ? Briefly describe
the operational principle of a time domain reflectometer.
6. With the help of a block diagram, describe the architecture of
a down conversion type microwave frequency counter. Com-
pare its features with those of a prescalar type microwave
frequency counter.
7. Briefly describe different techniques used in frequency syn-
thesisers. What are the advantages of using Direct Digital
Synthesis over other methods ?

SELF EVALUATION EXERCISE

Multichoice Questions
1. With reference to a two port network, with its input port
stimulated by an RF signal, if the magnitude of reflection
co-efficient is zero, then the return loss in dB would be
( a )zero ( b )infinity ( c ) 10 ( d )none of these
2. A three port microwave network can be characterised by a
scattering matrix having
( a )9 elements ( b )3 elemenm
( c ) 4 elements ( d )6 elements
3. Reflection co-efficient is
( a ) a scalar quality
( b ) a vector quality
( c ) independent of impedance mismatch
(d) one parameter that can not be expressed by any of scat-
tering panneters.
4. One of the following microwave power measuring set-ups
would usually be used for measurement of medium power
levels in the range of 10 mW to 1W
392 MICROWAVES AND RADAR

( a ) Using bolometer sensing element and making the

microwave power directly fall on the sensing element
( b ) Using a calorimeter
( c ) Using a directional coupler together with a low level
power meter
( d ) Using a calibrated attenuator together with a low level
power meter
5. A reflection co-efficient of 0.1 means a return loss of
( a ) 20 dB ( b ) 0 dB ( c ) 10 dB ( d )infinity
6 . A 20 dB attenuator would have the scattering parameter
(S21) magnitude of
( a )10 (b) 2 ( c ) 0.1 ( d ) 0.01
7. One of the following instruments can be used for making
transmission and reflection measurements
(a)Network Analyser ( b ) Spectrum Analyser
(c) Time domain Reflectometer ( d )Reflectometer
8. A receiver's noise temperature is measured by using Y-factor
method and hotkold standards. If Y = 2.5 and the tempera-
tures of hot and cold standards are 295 " K and 77" K.
respectively, then the receiver's noise temperature is ap-
proximately
(a)68" K ( b ) 109" K ( c ) 218' K (d) 191" K
9. Out of the following microwave power sensing devices, the
fastest response device is the
( a )Bolometer ( b )Thermistor
( c )Thermocouple ( d )Schottky diode
10. If X' is calibration factor of the power measuring set-up, then
the unknown RF power can be computed from :
( a )K (DC substituted power)
( b ) (DC substituted power)/K
( c ) K 1 (DC substituted power)
( d ) K~(DC substituted powert)
 RADAR
FUNDAMENTALS
adar is an acronym for Radio Detection And Ranging which

R tends to suggest that it is a piece of equipment that can be used

to detect and locate a target. Though this does continue to be
the primary function of a radar, a modern radar does much
more than just detection and ranging. It is used to determine the
velocity of moving targets and also find many more characteristics of
the target such as its size, shape and other physical features including
the type and number of engines used on an aircraft, for example. Radars
are extensively used in many civilian and military applications. This
chapter gives a comprehensive treatment to radar fundamentals, types
and applications. Radar antennas have been deaIt with at length in the
chapter on 'Antennas' and so it will be not covered here.
The topics covered in this chapter include :
What is radar Pulse doppler
Basic radar sys- radar
tem True and Ap-
Basic radar func- parent ranges
tions Tracking radar
Accuracy and Range tracking
Resolution Velocity tracking
Radar Range Pulse compres-
equation sion radar
Radar waveforms Synthetic aper-
Radar transmit- ture radar
ters Over-the-horizon
Radar receivers radar
Signal processor Monostatic and
Radar displays Bistatic radar
Radar classifica- Surveillance
tion radar
CW radar Laser radar
FM-CWradar Radar uidance
Moving target in- f
of missi es.
dicator
394 MICROWAVES AND RADAR

9.1. What is a Radar !

As outlined above, radar is an acronym for radio detection and
ranging. It is a stand alone active system having its own transmitter
and receiver that is used for detecting the presence and finding the exact
location of a far-off target. It does so by transmitting electromagnetic
energy, in the form of short bursts in most of the cases, and then
detecting the echo signal returned by the target. The ranging is com-
puted from the time that elapses between the transmission of energy
and reception of echo (Fig. 9.1). The location of the target can be
determined from the angle 1 direction of arrival of the echo signal by
using a scanning antenna preferably transmitting a very narrow width
Transm~tted
pulse
Recewed
pulse
Radar
System

t-
Fig. 9.1.
beam. As mentioned in the introductory paragraph, a radar today does
much more than just detecting a target and finding its location. The
radar can be used to determine the velocity of a moving target, track
the target and even determine some of the physical features of the
target. Of course, not a single radar type can be used to do all the
functions. There are differenttypes best suited to different applications.
In addition, a radar is a principal source of navigational aid to aircrafts
and ships. It forms a vital part of an overall weapon guidance or a fire
control system. Behind most of the radar functions lies its capability to
detect a target, find its range and determine its velocity.
9.2. Basic Radar System
The basic components of a radar system are shown in the block
schematic arrangement of Fig. 9.2. The radar signal waveform as
generated by the waveform generator modulates a high frequency
carrier and the modulated signal is raised to the desired power level in

I Receiver H Processor H ""'

Processor

Fig. 9.2.
RADAR FUNDAMENTALS 395

the transmitter portion. The transmitter could be a power amplifier

employing any of the microwave tube amplifiers such as Klystron,
Traveling Wave Tube (TWT),Crossed Field Amplifier (CFA) or even a
solidstate device. Though a magnetron power oscillbtor could also be
used but a power amplifier such as a Klystron is preferred because of
its high average output power delivery capability. The radar waveform
is generated at a low power level which makes it far more easy to
generate different types ofwaveforms required for different radars. The
most common radar waveform is a repetitive train of short pulses. CW
is employed to determine the radial velocity of the moving target from
doppler frequency shift, FM-CW is used where it is desired to measure
range with a CW waveform. Pulse compression waveforms are used
when it is desired to have the advantage of longer range capability of a
long pulse and higher range resolution capability of a shorter pulse.
These waveforms are shown in Fig. 9.3.

Gated C W

Modulated C W

Transmit Rece~ve
Pulse Compression Radar Waveform
Fig. 9.3
The average output power requirement of a radar could be as small
as a few tens of milliwatts for very short range radars to several
megawatts for Over-The-Horizon-Radar (OTHR).
The Duplexer allows the same antenna to be used for both trans-
mission as well as reception. It acts as a switch disconnecting the
receiver from the antenna during the time the relatively much higher
power transmitter is ON to protect the receiver from getting damaged.
On receive, the weak received signal is routed to the receiver by the
duplexer. The duplexer usually makes use of gas filled Trans-
396 MICROWAVES AND RADAR

mitmeceive tubes which are basically sections of transnlission line filled

'with a low breakdown voltage gas (Fig. 9 . 4 ~ )These
. tubes get fired due
to presence of high power to direct the transmitter output to antenna.
After the transmitter signal is radiated, these tubes de-ionise or recover
quickly to direct any received signal to the receiver input. A circulator
is sometimes used to provide further isolation between transmitter and
receiver (Fig. 9.46). A circulator as a component can also be used as a
duplexer (Fig. 9.4~). The circulator duplexer contains a high power RF
circulator comprising of signal couplers and phase shifters such that a
signal enteringone port has a low attenuation path only to the next port
in a particular direction. The low attenuation paths in the circulator
shown in Fig. 9 . 4 ~
are 1-2, 2-3, 3-4, and 4-1. All other paths are high
attenuation paths.

Antenna

Fig. 9.4
RADAR FUNDAMENTALS 397

The antenna acts as an interface between the radar transmitter

output and free space. Mechanically steered parabolic reflector anten-
nas and electronically steered antenna arrays are con~nlonlyused.
These have been discussed at length in chapter-8.
The echo signal received by the antenna is directed to the receiver
input. The receiver is usually of the superheterodyne type. The receiver
filters out-of-band interference. It also amplifies the desired signal to a
level adequate for operating subsequent sections.
The purpose ofsignal Processing is to reject the undesired signals
such as clutter and enhance the desired signals due to the targets. It is
done prior to the section that makes the decision as to whether the
target is present and in case of target being present, extracts the
information such as range, doppler etc.
Data Processing refers to the processing done after the detection
decision has been made. Functions like automatic tracking, target
recognition are examples of data processingin a radar system.
The Displays put the processing information in a form usable by
radar operators and others wanting to use the information such as Air
Traffic Controllers, Weapon System Operators and so on. The operation
of a radar and the sequence of events that take place from start to finish
can be summarised in case of a typical pulsed radar as follows :
The eansmitter generates a repetitive pulse train with each pulse
having a burst of RF signal. The pulse parameters, of course, vary with
the type of radar and the mode in which it is operating. The duplexer
routes the pulsed electromagnetic energy to the transmitting antenna
which concentrates the energy fed to its input into a narrow beam in
the direction of the intended target. At the same time, a time base is
initiated coincidingwith the transmission time instant of the pulse. The
electromagnetic wave propagates through the atmosphere. This wave
gets reflected from the target due difference in the impedance charac-
teristics of the targets. The impedance offered by the atmosphere (or
more precisely the free space) to the propagating electromagnetic wave
is 377 ohms and any discontinuity encountered causes the wave to get
reflected. The amount of reflection depends upon the characteristics of
the target. The target reflects the wave in all directions and the portion
of the reflected energy travelingin the direction of the radar constitutes
the echo or the backscatter. It may be mentioned here that the reflection
also occurs from ground and sea surfaces, atmospheric conditions like
clouds, turbulence etc. These reflections occurring in the direction of
the radar constitute clutter. The backscatter energy travels back to the
radar and a portion ofit along with a portion of the clutter is intercepted
by the radar's receiving antenna which in the present case is same as
the transmitting antenna. The amount of the backscatter energy inter-
cepted by the antenna depends upon the capture area of the antenna.
The received signal is routed to the receiver by the duplexer. The signal
which contains both the desired echo as well as the interfering signals
and noise gets processed in the receiver. The processed information is
398 MICROWAVES AND RADAR

then subjected to the detection threshold comparison and if the signal

is larger than the detection threshold, a detection is said to occur. If the
detection is caused by the desired target, a target is said to be present
and if the same occurs due to interfwing signals, detection is a false
alarm. The detection threshold in f a d is so chosen as to minimise the
probability of false alarm. Ailother dett\ction error is the one when the
radar fails to detect an existent target due to the target echo signal
being weak and being not able to crms the detection threshold. When
a detection occurs i.e. when the processed signal corsses the detection
threshold, the time base initiated at the start is strobed and the round
trip propagation time. rfieas~redto determine the target range. The
antenna's position encoders are also strobed to determine the angle of
arrival of the echo at the time of detection. If the target is a mobile one,
its radial velocity information is contained in the doppler shift which
can be made use of to determine the target velocity.
9.3. Basic Radar Functions
The basic functions that a radar can do include target detection,
identifying target location in range and angular position and determin-
ing target velocity. The radar performs these tasks provided that the
target echo signals after signal processing are sufficiently stronger than
the interfering signals like noise generatedin the receiver, clutter which
is unwanted signal echo due to reflections from land, sea, clouds etc.,
jamming signai-which is an intentional interference, (EMI) which is an
accidental interference from frierrdly sources such as communication
systems, other radars and spill&r which is due to leakage from
transmitter into receiver occurring mainly in the CW radars. It may
also be mentioned that not all radars are capable of measuring all the
above listed parameters.
Detection
Detection is the process of determining whether or not a target is
present. There are four possible conditions of detection. If a target is
present and a detection also occurs, the result is considered as correct
(Fig. 9 . 5 ~ )Similarly,
. if there is no target and radar display also shows
no detectioncondition, the result is again correct (Fig. 9.5b). Ifthe target
is present and radar fails to show it on display, an error is said to occur
(Fig. 9.512).But if the target is absent and radar shows detection, it is a
different form of error referred to as false alarm (Fig. 9.5d). Both of the
last conditions are error conditions, the one of false alarm is usually
considered far more serious and undesirable. Such a tricky situation
usually occurs because the target echo and interference signals have
more or less the same shape after they have been processed in the
receiver and only way to discriminate between the two is by amplitude
compari'son. The radar can often confuse between a weak target and a
strong interference residue. For this reason, detection can only be
described by probabilities, probability of detectioit and probability of
false alarm.
RADAR FUNDAMENTALS 399

Probability of detection (Pd) is the probability that for any given

evaluation of signal plus interference and a threshold, the result will
be a detection if the target is present. Probability of false alarm (Pfn)
is
the probability that interference alone will exceed the threshold for a
compound test.
There are many other terms related to the two probabilities. These
include False Alarm Number (FNA), Probability of Signal (Ps), Prob-
ability of Noise (Pn), False Alarm Time (FAT) and False Alarm Rate
(FAR). False Alarm Number (FAN) is the reciprocal of probability of

'
4i
3 ----- - ---. --_-- - . -_-.~?!les"o'd
.------
Target Echo Target Echo
Clutter (Range)

E Clutter
4
A _ t (Range)
(b)

at
u /Threshold
3 -------- -----------------------------

Fig. 9.5
false alarm. Probability ofsignal (Ps)is the probability that on any given
single test of signal plus interference and threshold, the result will be
a threshold crossing if the target is present. Probability of detection
(Pd) differs from probability of signal (Ps)in the sense t h a t the former
is the result of many consecutive signal echoes processed together.
Probability of noise (Pn)is the probability that noise alone will cross the
threshold for a single test. Probability of false alarm (Pfa)is related to
the Probability of noise in the same way as 'Pd' is to 'Ps'. False Alarm
Time (FAT) is the mean time between noise threshold crossings.
Pfa = l/(FAT). B
where B = System bandwidth a t the point of test
400 MICROWAVES AND RADAR

False Alarm Rate (FAR)is the average number of false alarms per
seconds and is given by reciprocal of False Alarm Time (FAT).
Also, FAR = Pf, . B
Target Location
The target location is expressed in terms of its range, azimuth
angle and elevation angle (Fig. 9.6). Range is the shortest distance of
the target from the radar regardless of direction. Azimuth angle is the
angle between the antenna beam's projection on the local horizontal
and some reference. The azimuth reference in case ofland based radars
is usually the true north. Ship borne radars are usually referred to ship's
head, which is a line parallel to the ship's roll axis. Amborne radars
reference the roll axis on the local horizontal plane. Elevation angle is
the angle between radar antenna's beam axis and the local horizontal.
Local horizontal in case of land based radars is the plane passing
through antenna's center of radiation and perpendicular to earth's to
earth's radius passing through the same point. For shipboard and
airborne radars, it is also the plane containing vehicle's pitch and roll
axes.

Fig. 9.6.
Ranging
Ranging is based on the principle of measuring the time delay
between the transmission of a pulse of electromagnetic energy by the
radar and the detection of the received echo (Fig. 9.7). The product of

@
.a
Radar

Transmit
Pulse
r - 7 Received
f j echo
~ o u n dtrip Propagation ~ i m e - - - 4

Fig. 9.7
RADAR FUNDAMENTALS 40 1

measured time difference and the velocity of propagation of electromag-

netic waves gives twice the target range. Time difference is measured
from the center of the transmitted pulse to the center of the received
echo in what is known as the centroid ranging. In another case used
less often and called leading edge ranging, it is measured from leading
edge of the transmitted pulse to the leading edge of the received echo
pulse. The target range can be expressed as :
R = c.t/2
where R = Target range
t = Round trip propagation time
c = Velocity of propagation
Range Ambiguity
Range ambiguity occurs when multiple target positions produce
the same reported information and the radar is unable to distinquish
between the two in terms of range. It occurs when the received echees
are attributed to wrong transmitted pulses. Refer to Fig. 9 . 8 ~ .The
transmit pulses T i , T2,T3 etc. produce echoes E l , E2, E3 respectively.
Most radar systems, for a given received echo, take the last transmitted
pulse as the reference for computing the round trip propagation time
and hence the range. If there are no other transmissions between a
transmittedpulse and the correspondingreceived echo, this assumption
is correct and the target is called a Range Zone-1 target and is ranged
correctly. Now, refer to Fig. 9.8b. Here, the echo for a given transmitted
pulse is received after the next adjacent transmitted pulse has been
radiated with the result that the radar, on the assumption of the Range

Zone-1 target, measures the range ass 'Ra'and not 'R' leading to
ambiguity. This is Range Zone-2 target and this fact could be used to
determine the true range 'R'from apparent range 'Ra'.To sum up, any
target for which the range zone is not known is ambiguous in range.
Since radars initially assume the targets to be Range Zone-1 target,
402 MICROWAVES AND RADAR

there are no ambiguities when the round trip propagation time is less
than the time period of the transmitted pulse train.
Range Quantisation and Granularity
Range Granularity results from the digital nature of the radar's
signal processing, data processing and display sections. Due to this,
data occurs in discrete values. The smallest output data change is
termed as granularity. Such radars which handle target positions
digitally can only report a target's range in discrete steps. If the smallest
output data change corresponds to a range of 100 m, radar would give
range information in steps of 100 m. A target with a true range of say
5.23 km will be displayed as having a range of 5.2 km and the one with
a range of 5.27 km will be ranged as 5.3 km. It must not be confused
with range resolution which is the smallest range difference between
the closely located targets which the radar can detect as two separate
targets. Range resolution and other related topics are discussed in
detail in the next section.
Angular Position
Angular position is determined by measuring angular co-ordinates
i.e. azimuth and elevation angles. These can be determined from the
knowledge of the direction of arrival of the echo pulse or in other words,
the direction of antenna's main beam a t the time of detection. This, of
course assumes that the detection signal originated from the direction
of antenna's main beam. Apparent ambiguity can occur if the signal
enters through the side lobe.
Measurement of Target Velocity
When the target is moving with respect to radar, the frequency of
received echo is different from the transmitted frequency by an amount
equal to the doppler shift whose magnitude and sense are proportional
to the magnitude of the radial component of the relative velocity
between the target and the radar and the sense of this velocity com-
ponent. The doppler shift is positive or the received frequency is higher
when the target is moving towards the radar and negative or the
received frequency is lower when the target is moving away from the
radar. Target's radial velocityinformationis extractedfrom this doppler
frequency shift. Doppler shift 'fd' is given by :
fd = 2 Vr/h
where Vr = Radial velocity difference between target and radar
h = Operating wavelength
If '0' is the angle between the targetlradar velocity vector and the
radar antenna axis (Fig. 9.91, then
fd = [(2 Vr/h ) cos 81
Further, if 'Oh' and 'W are the horizontal and vertical angles
respectively between radar's axis and the target's velocity vector, then
COS 8 = COS Oh . COS 8"
RADAR FUNDAMENTALS 403

Fig. 9.9
Doppler Ambiguity
Doppler ambiguities exist in all those radar systems where the
signals are processed digitally including pulse doppler radars, digitally
processed CW radars and so on. In case of digital processing, the spectra
of transmit and receive signals are discrete in nature. In case of a radar
making use of doppler principle, the received spectrum will be offset in
frequency from the transmit frequency by doppler shift. For extracting
true information, i t is therefore necessary to know the transmit spectral
Iines that caused the doppler shifted received echo. Doppler ambiguity
will occur if it is not known as to which transmit spectral line caused a
particular spectral line. As the data is sampled at a rate equal to the
pulse repetition rzte of the transmit pulses and the Nyquist criterion
must be met with, the unambiguous doppler frequency for a given pulse
repetition frequency is given by :
+
Af = [PRF/2]
where Af = Doppler shift = 2 VJh
+
If the target's maximum radial velocity is V,, then the minimum
value of PRF which is unambiguous in both magnitude and sense of
velocity is given by :
[ PRF ] min = 4 Vr/h
If it is desired to avoid ambiguity only in doppler magnitude, then
[ PRF ] min = 2. Vr/h
It is indeed a practice in some pulse doppler radars which rely on
detections in mulitple PRFs during the time on target to resolve
ambiguity in doppler sense. To sum up, we can say that low PRF radars
are unambiguous in range but highly ambiguous in doppler whereas
high PRF radars are unambiguous in doppler but highly ambiguous in
range.
Doppler Granularity
Doppler granularity does the same to doppler shift information
what range granularity does to range information. It is the result of
target echoes being processed in a discrete set of bandwidths called
doppler bins. The result is that the information on doppler frequency
shift and hence the target velocity is also in discrete form. If each
404 MICROWAVES AND RADAR

doppler bin is 100 Hz wide, the doppler shic information will be

displayed in steps of 10QHz.
%4. Accuracy and Resolution
Accuracy aad Resdution are the two most important parameters
conaerning the radar sys'tem performance. While Accuracy tells about
the precision with which a radar can measure individpal parameters
like range, doppler, angular co-ordinates etc., Resolution defines the
ability of the radar to effectivelyhandle multiple targets. Again we could
talk about Range Resolution, DoppIer Resolution, Angular Resolution
and so on. The two parameters are discussed in detail in the following
paragraphs.
Accuracy
Accuracy of measurment of a given parameter is the deviation of
the measured value from its true value. For instance, in case of target
range, an accuracy off 10 m for a target at 1k m would mean that the
measured range could be anywhere between 990 m to 1019 m. Accuracy
of radar parameters is usually expressed in absolute terms. Accuracy
of different parameters such as Range, Doppler, Angular position etc.
would be expressed in Meters, Hz and Degrees respectively.
There are two types of errors which lead to measurement inac-
curacy. These are ( a )Bias errors and (b) Noise errors. Bias errors refer
to continuous deviation or offset from the true value and are usually
due to factors like equipment miscalibration, servo lags etc. Noise errors
refer to :andom uncertainties in the measured parameter and are
primarily due to the echo signal getting contaminated by noise and
other interfering signals. These errors are expressed as standard devia-
tion of random uncertainties.
Though, broadly speaking, the mechanisms that spoil the accuracy
in case of major radar parameters like range, angular position and
doppler are the same, there are one or two changes in each of them. The
factors affecting range accuracy, doppler accuracy and tracking ac-
curacy are briefly described in the following paragraphs.
Range Accuracy :Range Accuracy primarily depends upon two
factors namely the accuracy with which the radar determines the time
of arrival of the echo pulse and the accuracy of measurement of time
delay between the transmitted signal and the received echo. The error
in the measurement of time of arrival is caused primarily by noise and
other forms of interference contaminating the echo. The time of arrival
is measured at the centroid of the echo pulse, though in some applica-
tions, leading edge of the envelope is also used. Due to noise and
interference, the signal centroid occurs at a time other than that of the
actual center of the echo pulse. The range error '6R'due to a time-of-ar-
rival error of 'ST' can be computed from :

where Vp= Velocity of propagation

RADAR FUNDAMENTALS 405

The other predominant source of error affectingrange accuracy is

the time delay error caused by erroneous time bases and miscalibration
of systems.
Range error due to this factor is given by :

where 6Tp = total round trip propagation error.

Angular PositionAccuracy :Then angular accuracywith which
a radar can establish the angular co-ordinates of a target include the
antenna pattern beamwidth, echo signal to noise ratio, scintillation
(target amplitude fluctuations) and glint (target phase fluctuations)in
case of a rangefinding radar and in addition the servo system noise in
case of a trackingradar. A narrower beamwidth, higher SIN ratio, lower
scintillation and glint and reduced servo noise are definitely the desired
fecltures. Precise determination of angular co-ordinates of a target are
very vital to a tracking radar. So, with reference to a tracking radar,
Figs. 9.10(a)and ( b )show the angular error as a function of target range
in case of

Fig. 9.10
conical scan and lobing and monopulse tracking radars. Contribution
of various error contributing factors outlined above is also presented in
these figures. As is obvious from the given curves, a monopulse radar
is not affected by scintillation. Thgerror dependance on beam width
and S I N ratio is expressed by :
406 MICROWAVES AND RADAR

where 83 = 3 dB beamwidth
z = pulse width
B = equivalent noise bandwidth
S/N = power signal to noise ratio
Kl = constant whose value depends upon the
tracking technique used (wheter conical scan or
monopulse )
Angular error due to glint is given by :
68, = K2/R
where K2 = Constant whose value depends upon target span and glint
parameters.
Scintillation track errors refer to the errors caused by amplitude
fluctuations and thus only thoseradars that depend for their operation
on sequential target amplitude variations to generate tracking infor-
mation are affected by scintillation phenomenon. It is particularly
severe a t low elevation angles.
Angular factor that can produce angular error is multipath
propagation of signal i.e. signal is travelling from radar to target and
back following more than one propagation paths. One effective solution
to this problem is to use a very narrow elevation beam so as to prevent
reflection points on ground and water surfaces from being in the main
response of the antenna.
Resolution
Measurement resolution of a certain radar parameter is the ability
of the radar to measure that parameter for multiple targets. For
example, high range resolution in a radar would allow the same to
measure range of two closely located targets in range. It also refers to
the ability of a radar to detect multiple features on the same target. For
example different parts of an aircraft may return different doppler
information back to the radar due to some relative motion between them
and this doppler information could be used by the radar to detect certain
physical features of the radar. Radar targets could be resolved in four
dimensionsi.e. Range, Azimuth Cross range, Elevation cross range and
Doppler.
Range Resolution
Range resolution is the ability of the radar to detect multiple
targets separated in radial range but having the same angular position
(Fig. 9.11). It depends upon the processed pulse width which in other
words is indicative of RF signal bandwidth. Except for the radars which
use special techniques to alter the received pulse width during process-
ing as is the case with a pulse compression radar, the processed pulse
width is the same as the transmitted pulse width. Narrower the pulse
RADAR FUNDAMENTALS 407

Fig. 9.11
width, closer can the two targets be in range and still be detected as
two separate targets rather than one big target. This phenomenon is
illustrated in Fig. 9.12. Range resolution is expressed as :
AR=c~/2=c/2B .
where T = Processed pulse width
B = Bandwidth
Effective bandwidth of any processed signal is approximately
reciprocal of its pulse width.

Processed Pulse Targets Unresolved Targets Resolved

W~dth /I"

Processed Pulse Targets Unresolved Targets Resolved

Fig. 9.12
To get a feel of the range resolution, it would be worth stating here
that a l p wide processed echo pulse would mean a range resolution of
150 m i.e targets separated in range by a distance equal to or greater
than 150 m would get resolved.
As is clear from the above description, a higher range resolution
means a shorter pulse width which in a conventional pulsed radar would
also mean a shorter transmitted pulse width. We have also seen earlier
that a longer pulse is required for a higher ranging capability which
implies that a conventional pulsed radar can not achieve b0th.a higher
range as well as a higher range resolution capabilities as the require-
ments for the two are opposing. Since from the application viewpoint,
a higher range capability is inescapable, a longer transmitted pulse is
unavoidable. In that case, one of the methods to enhance resolution is
to compress the recceived pulse during processin$. This is implemented
in a pulse compression radar. The operational hechanism of a pulse
compression radar will be discussed in the latter part of the chapter.
Cross Range Resolution
Cross range resolution is the resolution in a diection perpendicular
to the antenna axis. We have the azimuth cross range and the elevation
cross range (Fig. 9.13).
 MICROWAVES AND RADAR

Radar

Target-2

Fig.9.13
The xoss range resolution depends upon the antenna beamwidth.
Ncrrower besmwidth has higher resolution and thus cau resolve more
closc?',yspaced targets in the cross range. This is illustrated in Fig. 9.14.
Cross range resolution is expressed as :
AR ( cross range ) = R. 83 = [ R. h/Deffl

Targets Unresolved Targets Resolved

in Cross Range in Cross Range

Fig. 9.14.
A higher cross range resolution would mean an antenna with a
much large aperture. A technique to enhance cross range resolution
even with an antenna of a practical size is used in a SyntheticAperture
Radar (SAR)where the anknna is moved across the target and the echo
signals received with the antenna in different positions. These signals
are processed in a way as to achieve the same result as would have been
achieved if the antenna had an aperture equal to the distance moved
by the antenna (Fig. 9.15).
Doppler Resolution
Doppler resolution is the ability of the radar to detect and differen-
tiate between the targets at the same range and angular position but
having different radial velocities. Higher doppler resolution is par-
ticularly useful in identifying physical characteristics of the target such
as an &craft by resolving the net target movement and movement of
some other parts of the targets such as rotating blades of the engine on
the basis of doppler,As the moving targets produce a doppler shift
proportional to their relative radial velocity with respect to the radar,
the doppler shifts produced by the two targets will be different. This
difference between the two dopplers, that can be resolved, is reciprocal
of the look or dwell time. For instance, if the look time is 1 sec, the
RADAR FUNDAMENTALS 409

targets producing a doppler difference of 1 Hz can only be resolved. On

the other hand, a look time of 10 sec will allow doppler difference as
small as 100 i,lHz to be resolved. Similarly, if the look time is 0.1 sec,
the targets with a radial velocity difference as much as to produce a
doppler difference of 10 Hz will only be resolved.
Wd = 1/T
This means that a longer look or dwell time would be required to
achieve a higher doppler resolution. This goes against the desirable
requirements ofrange processingwhere a smaller look time is preferred
so as to ensure that the target range does not change much as the range
information is being processed.

Fig. 9.15

Problem 9.1 :Apulsed radar with a PRF of 1 kHz receives a n echo

pulse exactly 0.15 ms after it transmits. What should be target range in
km. Also determine the maximum unambiguous range of the radar i f its
range processing circuitry assumes range zone-1 target location.
Solution :Round trip propagation time, t = 0.15 ms
Therefore, Range = c.t/2
= [3 x 10' x 0.15 x 103/2] = 0.225 x l o 5 m = 22.5 km
PRF = lkHz = 1000 HZ
Therefore maximum unambiguous range
= [c/2 x PRF]

Problem 9.2 :A car is moving towards a stationary CW doppler

radar transmitting at 10 GHz along the axis of the radar with a speed
of 108 kmlhr. Determine the doppler shift and the frequency of the
410 MICROWAVES AND RADAR
received signal. What would be the received signal frequency if the car
was moving away from the radar along the same axis ?
Solution : f = 10 GHz
h = [3 x 10'/10 x 10'1 = 0.03 m
Doppler shift, fd = 2 Vr/h = 2 x 30 /0.03 = 2000 Hz = 2kHz
Received frequency
= 10000000 + 2 = 10000002 kHz = 9.00002GHz
If the car is moving away from the radar, the received frequency
is given by :
1000000 - 2 = 9999998 kHz = 9.999998 GHz

Problem 9.3 : A pulse doppler radar emitting at 10 GHz has a

repetition rate of 2 kHz. Determine i f this radar is capable of measuring
the radial velocity of 0.5 Mach of a closing target aircraft without any
ambiguity. If not, what should be the desired pulse repetition rate to
achieve that ?Assume velocity of sound to be equal to 330 m l s .
Solution : f = 10 GHz

PRF = 2 kHz
Now, maximum unambiguous doppler shift = PRF/2 = 1kHz
Also, doppler shift, fa = 2 Vr/h which gives Vr (unambiguous)

Velocity of the aircraft = 0.5 Mach = 0.5 x 330 m / s = 165 mls which
means that the radar is not capable of determining unambiguously the
velocity of the approaching aircraft. If i t has to do so, then the desired
repetition rate can be determined as follows :
For Vr = 165 m/s, fa = 2 x 165/0.03 = 11000 Hz = 11kHz
Therefore, the desired PRF = 2 x 11= 22 kHz

Problem 9.4 : A pulsed radar is transmitting 1 pi pulses at a

repetition rate of 1 kHz. Assuming that no pulse compression technique
has been used while processing the echo pulses, determine whether two
targets separated in range by 500 m but having same angular position
can be resolved by this radar on the basis of range.
Solution :Transmitted pulse width = 1ys
Received pulse width = lps
Range resolution = c .2/2 = 3 x lo8 x 10-~/2= 150 rn
where z = Received pulse width
RADAR FUNDAMENTALS 411

This means that radar can resolve upto a n inter target separation
i n range of 150 m. Therefore, given radar will be able to resolve the
targets i n question.

Problem 9.5: Two aircrafts having same radial range of 3 k m and

elevation co-ordinates are separated i n azimuth by a distance of 100 m.
Determine the maximum permissible 3 d B beamwidth of the radar
emission if the two aircrafts are to be resolved.
Solution : Cross range resolution = R. 83. [ d l 8 0 1
where R = Radial range
03 = 3 dB beamwidth
83 = I180 x cross range resolution/x: x R ]
= [I80 x 100/3.14 x 30001 = 1.91"

Problem 9.6 : Two aircrafts are approaching a ground based

radar emitting at a wavelength of 3 cm. First aircraft is closing i n along
the radar axis with a velocity of 0.5Mach. The second aircraft is flying
with a velocity of400 NMI hr and its velocity vector is making a n angle
of 30"with radar axis. Determine the required minimum look or dwell
time if the two aircrafts are to be resolved i n doppler. Assume 1 N M l hr
= 1.85 k m I hr and velocity of sound = 330 m /s.
Solution : Velocity of the first aircraft = 0.5 Mach = 0.5 x 330
= 16 m / s
Doppler = 2 x 165/0.03 = 11 kHz
Radial velocity of the second aircraft =
[400 x 1.85 x 100013600] cos 30" 102.75 m/s
Doppler = 2 x 102.75/0.03 kHz
= 6.85

Doppler difference= 11 - 6.85 = 4.15 kHz

Therefore,required look time = [1/4.15]= 0.24 ms

Problem 9.7: The specification sheet of a certain radar says that

the maximum unambiguous range of the radar is [1000000 / 12.4 x
PRF] NM. What does the number 12.4 in the denominator signify or
represent ?
Solution : Rmax = [ 1000000/12.4 x PRF] N M
= [lOOOOOO x t / 1 2 .4] N M
where t = Round trip propagation time in seconds
The numerator now represents round trip propagation time in
microseconds.
412 MICROWAVES AND RADAR

Therefore, number 12.4 represents the units microseconds per

nautical miles (p/NM).
In other words, this means that the round trip propagation time
for one nautical mile range is 12.4 p which is equivalent to 6.66 ps for
one k m range.
9.5. Radar Range Equation
The radar range equation relates the radar's detection range to
various radar and target parameters. These parameters include the
Transmitted power, Transmit antenna gain, Radar cross-section of the
target, Receive antenna aperture, Minimum detectable power at the
receiver input and various loss factors. The range equation has been
derived from first principles step-by-step in the following paragraphs.
Brief description of different parameters entering the range equation
has also been given along with different steps of derivation of equation,
in particular emphasising the significanceof these parameters vis-a-vis
the maximum detection range of the radar.
The first step is to determine the effective radiated power in the
direction of the main beam of the radar's transmitting antenna.' It is
the product of power delivered to the antenna input and the antenna
gain which implies :
ERP = Pt . Gt
where Pt = Transmitter power at antenna input
Gt = Transmitting antenna gain
Antenna gain as discussed in detail in chapter4 is the effective-
ness with which it concentrates the power fed to its input in a preferred
direction. An isotropic radiator when fed with power equal to ERP will
produce the same power density at a given range as that produced by
a highly directional antenna when fed with a power equal to ERP
divided by its gain. As the emitted electromagnetic energy propagates
in the forward direction, its power density, which is the power per unit
area of beam cross-section goes on reducing. Now area of the beam at
any given range is the area of sphere whoseradius equals the range in
case of an isotropic radiator. In case of an antenna with gain, it is the
area of the sphere with radius equal to range divided by antenna gain.
The power density at a range 'Rt' can be computed from :
Power density at the target = [ E R P / ~ R R or~[ ~~ It / ( 4 d ? / ~ t ) l
= [Pt. ~ t / 4 R R ? lw/m2
we should remember that i f a power (Pt)fed to an antenna produces a
certain power density at a given range, then a power ERP (= Pt. Gt) fed
to the isotrpic radiator wouldproctuce same power density at that point.
A part of this energy is reflected from the target depending upon
reflection characteristics of the target. It may be mentioned here that
reflection of energy occurs if the energy propagatingin free space having
an impedance of 377 ohms encounters a discontinuity in impedance in
the form of target. The part of reflected energy travelling towards the
RADAR FUNDAMENTALS 413

radar is called backscatter and it is this backscatter that we a;e

interested in. There are various target related parameters which
together with the incident power density determine the backscatter:
These parameters are combined together and expressed in the form of
target's radar cross-section (a).
Radar cross-section of a target is defined as the ratio ofits effective
isotropically reflected power to the illumination or incident power
density. It may be remembered that only the power reflected towards
the radar is a part of RCS.
RCS (o)= 4%[ Reflected power per unit solid angle 1Incident power
per unit area 1
Radar cross-section is made up of three components namely the
area of the target, target reflectivity at the polarisation of the receiving
antenna of radar and directionality of reflected power. The target's
radar cross-sectionfluctuates in both amplitude and phase and the two
types of fluctuations are respectively known as Scintillation and Glint.
While scintillation causes variations in the received echo power, the
glint affects the wavefronts ecchoing from the target.
Coming back to the radar range equation, the effective power
reflected from the target in the direction of radar can be expressed as

where Prt = Backscatter

o = Target's radar cross-section
Prt, the backscatter, propagates towards the radar and the power
density at the radar's receiving antenna due to backscatter 'Pd; can be
expressed by

where Rr = Range from target to radar receiving antenna

In case of monostatic radar (a radar with co-located transmitter

and receiver), Rt = Rr = R .
Therefore,
Pt Gt o
Pdr = -
(4%)'R4
If 'Ae' is the effective aperture area of the antenna, then
Pt Gt d e
Received power, Pr =
( 4 7 ~R4
)~
Effective area and gain are interrelated by
414 MICROWAVES AND RADAR

where h = Operating wavelength

This gives :
P~G~GG~A~
Pr =
(4n13R4
where Gr = Gain of receive anenna
For common transmit and receive antenna, Gt = Gr = G
Pt G 2 (3 h2
Pr =
(4nj3 R4
The received signal to noise power ratio determines whether or not
sufficient signal is present to detect the target. The noise power at the
input to the receiver is given by :
Pn = KToBF
where K = Boltzmann constant (1.38 x W-,/OK)
To = Absolute temperature
B = Noise bandwidth
F = Noise figure
The (SIN) power ratio a t the receiver input can be written as :
S I N = Pr/Pn
or Pr = K T a F (SIN)
If 'Smin)is the minimum detectable signal power, then the maxi-
mum radar range can be expressed as : - .

Since

Therefore, (Rmax14=
4n h2 Smin
The minimum detectable signal 'Smin) is usually 10 to 20 dB
stronger than the noise at a point in the receiver where detection
decision is made. The minimum detectable signal can be exaressed as
a signal to noise ratio, requiredfor reliable detiction, times t i e receiver
noise. Also 'Smin'is a statistical quantity and must be described in terms
of probability of detection and probability of false alarm. The above
equation, which can be expressed in various forms, is called the radar
range equation. In addition to its utility for prediction of range, the
equation also forms a good basis for preliminary system design by
allowing the designer to appreciate the effect ofvarious radar and target
RADAR FUNDAMENTALS 415

parameters on range and thus optimise them for best performance. The
above equation however assumes that there are no propagation losses
as the signal propagates from radar to target and from target to radar
and also that the target lies in the beam maxima. The propagation
losses are due to absorption, diffraction and certain types of refraction
effects etc. The losses are expressed in the form of pattern propagation
factors 'Ft' and IF;.. 'Ft'
. is defined as the ratio of field strength 'E at the

target to that which would exist at the same distance from

radar in free space and in the direction of antenna beam maxima. 'F;
is similarly defined. This modifies the range equation to :

Another loss element is the loss of power in the transmission line

connecting the transmitter output to the transmitting antenna input.
In that case 'Pt' is to be replaced by [Pt/Lt]where 'Lt' is the ratio of
transmitter output power to the power actually delivered to the anten-
na. For a lossless transmission line, Lt = 1.There can be other similar
loss factors. All these loss factors are multiplicative and can be ex-
pressed as a single loss factor 'L'which gives another form of range
equation that takes into account the loss factor as :

Evaluation of Range Parameters

The factors which determine the range of a radar discussed in the
preceding paragraphs are highly interdependent. The definitions of
most of these factors are arbitrary. Therefore, while evaluating these
factors, it is important that the definitions chosen are mutually com-
patible.
'Piis the transmitted power called peak transmitted power. In
pulse radars, it is the average power during the pulse. If 'z' is the pulse
width and 'W(t)' the instantaneous power, then

A= p2
7 -2/2
W (t) dt where 7 = Pulse width

It is clear from the range equation that the transmitted will have
to be increased 16 times in case it is desired to double the range with
other parameters remaining unchanged. This means that it will not be
feasible to increase ranging capability of the radar by increasing the
transmitted power without any limit.
Antenna gain is the other parameter that can be used to enhance
the rangingcapability of the radar. If the same antenna is used for both
transmission a;ld reception, then
416 MICROWAVES AND RADAR

This shows that the antenna gain needs to be increased by a factor

of four only in order to double the maximum range.
Xadar cross-sectionof the target is another parameter that affects
the range. This is afactor that is beyond the control of the radar, though
it has been effectively utilised by airxafts to avoid detection by radar.
The modern fighter aircrafts are so designed as to minirnise their radar
cross-section by using what is known as stealth technology (discussed
in detail in chapter-11 on Electronic Warfare).
Presence of 'A' (or 'p) in the radar range equation makes it clear
that range equation is frequency dependent. The frequency dependence
of the range is a complex phenomenon. It is so because a number of
other parameters are dso frequency dependent. For example, most
antennas have frequency dependent gain.
The radar range is also affected by the receiver bandwidth and the
receiver signal to noise ratio. A more sensitive radar receiver would
have a higher ranging capability. To sclm up, it is not practical to play
with various range determining parameters outlined above without
considering their effects on the other riesired performance specifications
of the radar system. One also neede to weigh the effectiveness of a
certain parameter to enhance the range against its implications on
other performance specifications. H i g ~ e rantenna gain and a more
sensitive radar receiver, for instance, seem to be much better options
than increzslng the transmitted pofier to increase the range.
9.6. Radar Waveforms
A large variety of transmit waveforms are employed in radar
systems depending upon the type of radar and its intended function.
Waveforms other than the convmtional gated CW and the CW are often
used by the radar to enable it either perform additional functions or
perform the existing functions with enhanced capability. For example,
a CW radar capable of measuring target speeds only could be made
capable of measuring the target range also by modulating the trans-
mitted signal using frequency modulation such as in FM-CW radar.
Also, parameters like Range resolution, Doppler resolution could be
enhanced by using special waveforms. Waveform characteristics like
its spectrum, spectral bandwidth, auto-correlation function etc. can be
used to determine the suitability of a given waveform. For example, the
Fourier transform of the waveform's auto-correlation function deter-
mines its power spectrum. Also, narrow auto-correlationindicates good
range resolution capability. In the paragraphs to follow, we shall briefly
examine some commonly used radar waveforms along with their char-
acteristic features.
Gated CW Pulsed Waveform
Gated CW pulsed waveform is commonly used in pulsed radars. It
is basically a sinusoidal signal gated to a rectangular time envelope,
the frequency versus time plot of which is shown in Fig. 9.16. This type
of waveform can be expressed by the generic expression :
RADAR FUNDAMENTALS 417

[ u ( 0 )- u (TI].A cos [ + ( t )+ hl
where u ( 0 )and u ( T ) are unit step functions occurring at time t = 0 and
t = T respectively.
Also, A = Peak amplitude of the waveform
$(t)= Waveform's phase function
+, = Absolute phase of the wave usually defined as zero

Fig. 9.16
Now, if 'f,' is frequency of the sinusoidal signal in the gated pulse,
then the waveform phase can be computed from :

where o = Radian frequency = 2nfo

This gives +(t)= 2nfot
The frequency domain response is given by :
[n
sin (E- fo) T ] h ( f - f o ) T
where f, = Frequency of sinusoid being gated
T = Gating pulse width
The time domain and frequency domain responses are shown in
Figs. 9.17(a)and ( b ) .

Fig. 9.17
Now sin x / x =0 gives x = f t m
418 MICROWAVES AND RADAR

which implies that frequency domain response will have nulls located
at frequencies giveil by :

which gives f = Cfo +- n/xl

Therefore rlulls are located at values of 'f' which are away from
'fo' by a quantum equal to integral multiple of reciprocal of '3. The main
lobe has a width equal to '2/.2. The matched bandwidth equals 'l/3.
Different spectral lines (each line representing an individual sinusoid)
are separated by the frequency equal to PRF.
Linear Frequency Modulated (LFM) Gated Pulse
In the LFM gated pulse also known as 'Chirp', the frequency is
swept linearly across the pulse. Frequency versus time graph is as
shown in Fig. 9.18(a).The graph shown is for a Down-Chirp where the
frequency is swept from a higher to a lower value. In the Up-Chirp
shown in Fig. 9.18 (b), the frequency is swept from a lower to a higher

Fig. 9.18
value. The advantage of having modulation in the transmitted pulse is
that it is possible to achievelarge bandwidth even with long transmitted
pulse widths. In an unmodulated pulse discussed earlier, the bandwidth
is approximately equal to reciprocal of the pulse width and the only way
to enhance bandwidth is by reducing the pulse width. The chirp and
other modulated waveforms achieve their bandwidth through modula-
tion and not the pulse width. If the frequency is swept from 'fi' to 'fz',
then
dfldt = fi - f i ) / ~
fi > f2 for Down Chirp and fi c f2 for Up-Chirp
Instantaneous frequencyfit) as a function of time is given by :
At) = [(fz-fi)/rl . dt = [V2 - f i ) / ~ l . t +K

At) = [(fi - fWr1. t + fl

Phase function + (7) can be expressed as :
444 ldICROWAVES AND RADAR

sinusoidal or a triangular modulating signal. Another way of achieving

ranging capability with a CW radars by using phase coding. The
n~axinlumrange capability on all the three cases above would be
determined by period of n~odulation.
FM-CW R a d a r With Sinusoidal Modulation
The modulating signal here is a sinusoidal signal which leads to a
sinusoidal frequency versus time characteristics of the transmitted
signal. The range is determined by the instantaneous phase shift
between the transnut and receive signals. Fig. 9.50 ( a ) shows the
Frequency versus Time characteristics of the transmit and echo signals
when there is no doppler shift. The phase shift 'Af' corresponds to a two
way propagation time of 'Tp'given by :
Tp = [A$/360 . fm]
where A$ = Phase difference in degrees
fm = Modulating frequency

Modulation

Transmit
Modulation
7

,Echo Modulation

Fig. 9.50
If the target is moving, the echo signal will also have a doppler
shift (Fig. 9.50b). The doppler shift 'fd' is indicative target's radial
velocity. Doppler is extracted by comparing DC values of the demodu-
lated transmitted and echo signals. Sinusoidd FM-CW radar has a
disadvantage that it is not capable ofresolving multiple targets. If there
are more than one targets, each demodulated echo will be sinusoid of
modulatingfrequency having a phase shift and DC offset corresponding
to its own target range and velocity. The demodulated echo from
multiple targets is then the sum of demodulated echoes from individual
targets with the net result that the system perceives it is a single target
with range and doppler which is an average of all. Sinusoidal FM is
therefore used only in case of single target track radars.
RADAR FUNDAMENTALS 445

FM-CW Radar With Triangular Modulation

The transmitted signal in this case is modulated by a triangular
signal which leads to an up sweep portion where the frequency increases

Fig. 9.51.
to'a high value 'fd and a down sweep portion where
from alow value ' f ~
the frequency decreases from a high value 'fd back to a low value ' f ~ ' .
Fig. 9.51 shows the Frequency versus Time characteritics. In case of an
FM-CW radar with triangular modulation, i t is the instantaneous
frequency difference between the trznsmitted signal and the received
echo that determines the range. Fig. 9.52 shows the Frequency versus
Time characteristics of transmit and echo signals in case of a target
producing a zero doppler shift. The instantaneous frequency difference

Echo Modulation

I
4-1 _C

&T+ t
Fig. 9.52
in this case is constant throughout the up-sweep and down-sweep with
the difference that transmit frequency is greater during the up sweep
and it is the opposite during the down sweep. The range can be
computed from :

where B = f~ - f~
Af= Instantaneous frequency difference
T = Round trip propagation time
In case the target produces a doppler shift 'fd', then the instan-
taneous frequency difference during the up-sweep and down-sweep is
respectively given by
[- Af + f l [Af + ff;.
~ and
446 MICROWAVES AND RADAR

Af = (Received signal frequency - Transmit signal frequency) due

to range delay only
NOW, f (US) = - Af+ fd
and f (DS) = Af + fd which gives
fd = CjC(DS) + f (US)]/2 and Af = [f(DS) - f (US)]/2
Thus both doppler and range information can be determined by
measuring instantaneous frequency difference during up sweep and
down sweep. While using the above expressions, due consideration
should be given to the sign of the instantaneous frequencies i.e. up-
sweep difference frequency [f (US)]and down-sweepdifference frequen-
cy Lf(DS)]with the former to be given a negative sign and the latter
obviously the positive as the two have opposite sense. To simplify the
calculations, if we only take the magnitudes of the two difference
frequencies, then the average of the two gives 'Af and half of the
difference between the two gives fd.
Such a radar is also capable of resolving more than one targets. In
this case, individual targets produce their own-up-sweep and down-
sweep frequency differences which can be sorted out by spectrum
analysis of the composite received signal.
Since doppler shift is a function of the transmit frequency, the
doppler shift may itself change significantly during sweep period. In
order to avoid this problem, sweep bandwidths in most FM-CW radars
are much narrow as compared to carrier frequency. Typically, it is a few
MHz for a carrier frequency of several GHz.
Phase Coded FM-CW Radar
The transmit signal is CW with phase modulation. Range in this
case is given by the time difference between transmission of a code
segment and reception of the same (Fig. 9.53). The time difference is
measured using correlation technique. The received signal is correlated
with delayed transmit signal. The delay that corresponds to correlation
function maximum is the propagation time. Range inambiguity in this

j ;--tp-----4 'lransmit Wave

1 Received Echo
--- ----
---
Fig. 9.63
case requires that the selected code has only one peak in its auto-cor-
relation function over the range of possible propagation times.

Problem 9.18:AnX-band single target track radar uses a transmit

signal that is a carrier of 10 GHz modulated by a 100 Hz sinusoidal
sig~zal.The peaks of the modulating signal (corresponding to the maxi-
RADAR FUNDAMENTALS 447

m u m frequency of transmitted FM sig~zal)and the demodulated echo

signal are l o o apart. Determine the target range.
Solution : Target range 'R' can be written as :
R = c . T p / 2 where T p = Round trip propagation time
NOW,Tp = A @ I(36O fm)
where A@= Phase difference between transmit and receive modulatioi~s
in degrees
fm = 100 HZ,A$= 10"
Therefore, Tp = 10/36C x 100 = (1/3600)s

Problem 9.19 : A n FM-CW radar using triangular modulation

and operating at a center frequency of 10 GHzproduces an up-sweep and
down-sweep frequency differences between the transmit and receive
signals of 60 kHz and 80 kHz respectively. Determine the target range
and radial velocity in Nautical Miles per hour if the modulation frequen-
cy is 100 Hz and the sweep bandwidth is 2 MHz.
Solution :AUS) = 60 kHz, f ( D S )= 80 kHz
Therefore, fd = (80 - 6 0 ) / 2 = 10 kHz
Af(range) = (80 + 6 0 ) / 2 = 70 kHz
Therefore, Range, R = c x T x Af/2 x B
=3 10' 5 10-3 70 103/2 2 lo6
= 26.25 km.
If 'V,' is the target radial velocity and 'f the operating frequency,
then
fd = 2.Vr.f/c

- -- -- - - -

Problem 9.20 : For problem- 19, determine the maximum error

that can occur in the measurement of abppler shift due to carrier
frequency sweep.
Solution :fd = 2.Vr/h = 2.Vr.f/c
For 'Vr'and 'c' as constant (for a given radial velocity, V, is
constant),
where 'j'is the operating frequency and K = 2. V r / c .
fd = K.f
Therefore, 6fd = K. 6f
Here, S f = 1 MHz around the center frequency
_+
 Therefore, 6fd =f 10-6 x 106 = f 1 Hz

Problem 9.21 : Refer to the three detection situations encountered

in an FM-CW radar and shown in -7ig..9.54 (a), (b) and (c). For each of
them, determine the range and the magnitude and direction of target$
radial velocity component. Assume center of operating frequency as 10
GHz.
Solution: Fig. 9.54 ( a ):The up-sweep and down-sweepfrequency
differences are the same. Therefore, the target is producing zero doppler
shift which implies that its radial velocity component is also zero.
The instantaneous frequency difference due to range is given by :

r&\&
10MHz
L \"/it '.JO '"/

1G MYz
#'

Fig. 9.54
Af= (100 + 100)/2 = 100 kHz
Therefore range 'R' can be computed from :
R=c.T~if/2.B
where T = Up-sweep period = Down sweep period = 5 ms
B = Sweep bandwidth = 10 MHz
~ = 3 x l 0 ~ x l5o 3xx 100x l o 3 / 2 x l o x 1 0 ~ = 7 5 0 0 m = 7 . 5 k r n
Fig. 9.54 ( b ): Doppler shift, fd = (80 - 50)/2 = 15 kHz
Frequency difference due to range, Af = (80 + 50)/2 = 65 kHz
RADAR FUNDAMENTALS 449

Range, R = c. T A f / 2 . B = 3 x 10' x 10 x x 65 x lo3/2 x 10 x lo6

= 9750 111 = 9.75 km
The radial velocity 'V,' can be computed from :
Vr= c. fd/2.f where f = operating frequency = 10 GHz
V, = 3 x 10' x 15 x 1 0 9 2 x 10 x lo9 = 225 m/s
As the up-sweep frequency difference is less than the down-sweep
frequency difference, this implies that the doppler shift is contributing
towards an increase in the echo signal frequency. Therefore, the target
is moving towards the radar.
Fig. 9.54 (c) : The doppler shift, fd = (80 - 50)/2 = f 5 kHz (same as
that computedincase of Fig. 9.54 ( b ) .Frequency difference due to range,
Af= (80 + 50)/2 = 65 kHz (same as that computed in case of Rg. 9.54(c)
Therefore, Range, R = 9.75 km
Magnitude of radial velocity = 225 m/s
In this case, however, the up-sweep frequency difference is greater
than the down-sweep frequency difference which means that the dop-
pler shift is contributing towards a decrease in the echo signal frequen-
cy. Therefore, the target in this case is moving away from the radar.
9.14. Moving Target Indicator (MTI)
The Moving Target Indicator (MTI)radar system very effectively
handles moving targets such as aircraft and is capable of measuring
their range and radial velocity component in the presence of strong
clutter due to stationary and even slow moving undesired objects such
as buildings, clouds, rain etc. It is again based on the doppler shift
imparted to the transmit signal by the moving target to determine the
target's radial velocity component. The range, of course, is measured
from the time lapse between the transmit signal and the received echo.
The doppler shift is not measured exactly in the same way as it is in
case of a CW radar where the process is more or less straightforward.
An MTI, being a pulse system, relies on the phase difference between
the transmitted signal and the corresponding echo to compute the
doppler. This phase difference for successive transmit pulses of RF
energy and their corresponding echoes changes in case of moving
targets at a rate equal to the doppler frequency shift. The phase
difference however remains the same in case of stationary targets and
changes a t a very small rate in case of slow moving targets so as to be
easily distinguishable from the phase difference information produced
by relatively much faster desired targets. The principle of operation of
echo signal processingis shown in the block schematic of Fig. 9.55. Each
echo from a given range gate is subtracted coherently from a delayed
version of the previous echo from the range gate. If the target is
stationary, both echoes would produce the same phase difference and
there would be complete cancellation provided the noise is absent. If
the echo has changed phase slightly due to target motion, the cancella-
430 MICROWAVES AND RADAR

tion would be incomplete. For a target in uniform motion, there would

be a constant change in phase from pulse to pulse and there is no
cancellation.
D~fference
Phase Delay 1 Pulse
Senslt~ve
From I F Detector Interval
section O/P k

from Local
osc.
- 9.55
Fia.
As is clear from the above description, there is a need to maintain
phase coherence because it is in the phase difference between the
transmit signal and received echo for successive pulses where the
target's doppler information resides. Based on the mmner in which this
phase coherence is ensured, there are two commonly used MTI con-
figurations. These are Coherent MTI and COHO-STALOtype MTI. The
latter is basically a coherent-on-receivesystem. The basics of these two
configurations have been discussed in an earlier part of the chapter.
Fig. 9.56 shows the block schematic arrangement of a Coherent MTI
system. Here, the transmitter is a power amplifier with its RF source
controlled by a crystal reference. Both the local oscillator as well
reference oscillator frequencies are synthesised from a stable reference.
The transmitted frequency is the sum of the two produced by Mixer-1.
Power
-
Duplexer
Amplifier
4 I

Mixer STALO Mixer

Phase e
COHO Detector
01p
Fig. 9.56
The reference oscillator output is also at the receiver intermediate
frequency (IF). The received signal is routed to Mixer-2 whose other
input is from the local oscillator. The mixer output at the receiver IF,
which is same as the reference oscillator frequency, is fed to the phase
sensitive detector where it is phase sensitively detected. The output is
fed to the signal processor section similar to the one shown in Fig. 9.55
earlier.
In the COHO-STALO system shown in Fig. 9.57, the transmitter
is a power oscillator, usually a magnetron oscillator. In this case, the
receiver has an extremely stable local oscillatorcalled STALO.A sample
of the transmitted RF output power at low levels is mixed with STALO
RADAR FUNDAMENTALS 461

Power
Oscillator Duplexer
I I

Mixer STA LO Mixer

+ +
COHO - Phase '
Detector 01p
e

Fig. 9.57.
output and lockingpulse thus generated then triggers another oscillator
called COHO whose output becomes the coherent reference. It is clear
that it is difference output of Mixer-1here that triggers the COHO. Rest
of the system is self explanatory.

Transmit Spectrum ( f T )
C

Echo Spectrum ( f R )

m
6fd f
(a)

Fig. 9.58.
452 MICROWAkES AND RADAR

Blind Speeds in MTI

As mentioned earlier, !he computation ol doppler shift ir case of a
pulsed radar is not as simple and straightforward as it is in case of a
CW radar. In a CW radar, doppler shift is recovered by measuring the
frequency displacement of the echo spectrum from the transmit
spectrum (Fig. 9.58a). The process is simple because the transmit
spectrum is a single line. In pulsed systems, the transmit spectrum
comprises of an infinite number of spectral lines separated by the PRF.
In such a case, if the target does not produce any doppler shift, the
situation is shown in Fig. 9.58(b). This is the case when the target is
either stationary or is moving at the same rate as the radar or its
velocity vector is perpendicular to the radar axis. The echo and the
transmit signal are at the same frequency. If the target's radial com-
ponent of velocity is such that the doppler shift produced is less than
half of the PRF, the Nyquist criterion is satisfied and the doppler
information is extracted without any ambiguity. The situation 1s
depicted in Fig. 9.58(c). It may be mentioned here that frequency of a
sampled wave is recovered as the smallest frequency span from the
received spectrum line to the closest transmit spectral line. The sam-
plingrateis the same as the PRF. How, if the target's radial component
of velocitfr vector is such that the doppler shift produced as a result is
more than half of the sampling rate i.e. PRF, as shown in Fig. 9.58(d),
the doppler measurement is now ambiguous as the doppler measure-
ment still reads the doppler shift as the location of the received spectral
line to the nearest transmit line. The ambiguity in doppler measure-
ment arising out of undersampling also called aliasing leads to the
reported or apparent doppler shift being different from the true doppler
shift. The true and apparent doppler shifts are interrelated by :
f~ = [(fd MOD PRF) - PRF] MOD PRF
or f A = [(fd MOD PRF) + PRF] MOD PRF
whichever gives a smaller absolute value.
The MOD operator is the remainder of the division process. For
instance, A MOD B is the remainder of the division of 'A' by 'B'. As an
illustration, for fd = 1000 Hz and PRF = 400 Hz, fd MOD PRF would be
203. Similarly, PRF MOD fd would be 400 as 40011000 gives 'a
remainder of 400 only.
Another very serious problem is that of Blind Dopplers or Blind
Speeds. When the doppler shift equals an integer multiple of PRF, the
moving target echo signal's spectral lines coincide with the spectral
lines of the transmit signal and so are the spectral lines of the stationary
target echoes (Fig. 9.58e). Another way of saying the same thing would
be that when the target's radial velocity component is such that it
travels a distance of'n.h/2' (where 'n' is an integer) along the radar axis
during the time between successive transmit pulses, then the phase
difference between the correspondingecho pulses would be '2n7c'radians
which is equivalent to no phase change or a stationary target. Though
use of doppler filters effectively attenuate echoes at zero doppler shifts
RADAR FUNDAMENTALS 453

and at integer multiples of PRF for clutter rejection, a moving targt:

producing these doppler shifts can not be detected. Such doppler shifts
and the associated radial velocity components are called BlindDopplers
and Blind Speeds respectively. The blind doppler shifts and blind speeds
can be computed from :
fr, = n.PRF
VB = [n.c. PRF/2f ]
where n = f 1,t2,f 3, f4.....
f = Operating frequency
One of the effective solutions to the problem of blind speeds lies in
PRF staggering. Blind doppler shifts are the doppler shifts equal to PRF
or its integer multiples. For a fixed PRF, the doppler phase is sampled
at the same point in each cycle. If the PRF was varied, the doppler phase
would be sampled at different points in different cycles with the result
that coincidence of doppler shift spectral lines with the transmit signal
spectral lines can be avoided thus facilitating recovery of doppler
information. The PRF stagger can be pulse-to-pulse o r look-to-look or
even scan-to-scan.

Problem 9.22 :A n MTI radar system operating at 10 GHz and a

repetition rate of 1000 Hz receives echoes from an aircraft that is
approaching the radar with a radial velocity component of 1 k m l s .
Determine the radial velocity component as measured by the radar.
Solution :True radial velocity of the target, V, = 1 k d s = 1000
mls
If 'fd' is the true doppler shift, then fd can be expressed as :
fd = 2 V,. f/c where ' f ' is the operating frequency

= 66.67 kHz
If ' f ~is' the appearant doppler shift, then
f i = [(fd MOD PRF) - PRF] MOD PRF
or f i = [(fd MOD PRF) + PRF] MOD PRF
whichever gives smaller absolute value
From the first equation :
f~ = [(66670 MOD 1000) - 10001 MOD 1000
= [670 - 10001 MOD 1000 = [- 3301 MOD 1000 = - 330 HZ
From the second equation :
f~ = [(66670 MOD 1000) + 10001 MOD 1000
= [670 + 101)0] MOD 1000 = 1670 MOD 1000 = 670 HZ
The first equation has yielded lower absolute value. Therefore, the
apparent doppler shift is - 330 Hz.
454 MICROWAVES AND RADAR

Now, radial velocity corresponding to this doppler shift is given

by :
vr= (Apparent)= ~ . f d / 2=f - 3 x lo8 x 330/2 x 10 x 10l0 = - 4.95 m/s
The radar measures the target to be moving away from the radar
at a radial velocity of 4.95 m/s though in reality it is moving towards
the radar with a velocity of 1000 mls.

Problem 9.23 :Verify that if the radar PRF is twice the expected
doppler shift, the radar would be able to measure the magnitude of the
dopplershift only without any ambiguity i.e. the measured doppler shift
magnitude will be the same as the actual doppler shift and also that it
would not be able to measure the doppler shift sense.
Solution : Let fd and f i be the true and apparent doppler shifts.
fd and fA are interrelated by :
fA = [(fd MOD PRF) - PRF] MOD PRF
or [fd MOD PRF) + PRF] MODPRF
whichever gives a smaller value
Considering the first alternative :
If PRF is greater than 'fd', then (fd MOD PRF) equals 'fd'
Therefore, f A = (fd - PRF) MOD PRF
Let PRF= 2 fd

Considering the second alternative :

f i = [3fdl MOD [2 fdl = fd
The two solutions have led to the same magnitude for the apparent
doppler shift. So, there is no question of preferring one over the other.
But the two solutions have given opposing sense for the doppler shift.
From this we can conclude that we have not been able to measure
unambiguously both the magnitude as well as sense of the doppler shift
here.

Problem 9.24 : Verify that if the PRF (i.e. sampling rate) is equal
to or greater than four times the true doppler shift, both magnitude as
well as sense of the doppler shift can be determined unambiguously.
Solution : Let PRF = n. fd where n > 2
and f A = Apparent doppler shift
fA = [(fd MOD PRF) - PRF] MOD PRF
or [(fd MOD PRF) + PRF] MOD PRF
whichever gives a smaller absolute value
fi = [(fd MOD n f d - nfd 1 MOD [nfdl
= Ifd ( 1-n)l MOD [nfd]= [- fd ( n - I)]MOD [nfd]= - ( n - 1)fd
RADAR FUNDAMENTALS 45s

Also f i = [(fd MOD PRF) + PRF] MOD PRF

= [(fd MOD nfd) + nfdl MOD nfd
= V d (1 + +)I MOD nfd = fd
The second alternative has yielded a smaller absolute value.
Therefore, f i = fd which is also the true doppler shift.
It is also clear from the above solutions that it is o h y for n > 2 that
apparent doppler shift is the true doppler shift in both magnitude as
well as sense.

Problem 9.25: Verifj,that the product of maximum unambiguous

radial velocity 'V;and maximum unambiguous range 'R'is independent
for radar's PRF and is equal to 'Kh' where 'I? is a constant.
Solution :Maximum unambiguous doppler shift is related to the
radar's PRF as :
fd = PRF/2
Since, fd = 2 V r / h
Therefore, 2 V r / h = PRF/2 which gives Vr = h . PRF/4 (11
Also, R = c / 2 . PRF (2)
From equations (1)and (2),
Vr.R = (c/8). h = K.h where 'K' is constant and equal to (c/8)
Problem9.26: A 3 cm MTIis operatingat a PRFof2000Hz. Find
the lowest blind speed.
Solution :Operating wavelength = 3 cm = 0.03 m
PRF = 2000 H z
Lowest blind speed = [nh/2]PRF for n = 1
Therefore, lowest blind speed
= [h/2].PRF = [0.03/21 x 2000 = 30 m / s = 108 h / h r

Problem 9.27: Two MTI radar systems are operating at the same
PRF but have dzrerent operating frequencies. If the third blind speed
of one is equal to fifth blind speed of the other, find the ratio of their
operating frequencies.
Solution : Let operating frequency of first radar = fl
and operating frequency of the second radar = f2
Now, third blind speed of first radar = [3.c.,'2 fi 1 . PRF
Also fifth blind speed of the second radar = [5.c/2. fil . PRF
Therefore, 13 c/2fil. PRF = [5c / 2 fi] . PRF
or [3/fil = [5/fil which gives [ f i / f i l =3 / 5
466 MICROWAVES AND RADAR

9.15.Pulse Doppler Radar

The principle of operation of a Pulse Doppler Radar is similar to
that of an MTI radar in the sense that both make use of doppler shift
caused by reflection of transmit signal from a moving target. Low PRF
radars designed to avoid rang? ambiguity problem and based on doppler
shift tci extract target speed information have become to be known as
MTI radars to distinguish them from pulse doppler radars which
usually hrp high PRF systems. Medium PRF pulse dopplars are also
prevalent. To establish connectivity with what we have already dis-
. ~ earlier pages, the main difficulty with MTI systems is that
c u s s e d ~the
the d )pplel: sampling rate, which equals the radar PRF, is too low for
the speed of the modern aircraft with the result that doppler shift
infwrna'ion from most of the realistic targets is highly undersampled.
Tk3 Nyqtlist criterion is not met and this essentialiy leads to aliasing
proh1e.r. Aliasingfurther leads to ambiguous estimates of target speed.
Occ ~rrenceof blind speeds, where the target appears stationary and
cnresolvable against background clutter, is still worse.
Pulse doppler radar, being a high PRF radar, offers solution to
such types of doppier ambiguities. The PRF is classified as high when
it is fast enough to sufficientiysan~ple the highest possible doppler shift
induced by the moving target. In fact, it is to be.?t least twice the doppler
shift. Assuming that targets with maximum closing velocities are to be
accounted for typical PRF for an X-band radar may be in the range of
300 kHz or so. Such a high pulse repetition rate however does not allow
unambiguous recovery of target range. The maximum unambiguous
range for a PRF of 300 kHz, foi- instance, would be 500 m only which
for all practical purpose would be as good as nothing.
Pulse doppler radars are generally divided into two broad PRF
categories namely medium PRF and high PRF. While a high PRF pulse
doppler radar, as mentioned above, is ambiguous in range and unam-
biguous in velocity, a medium PRF radar is ambiguous in both range
and velocity and these ambiguities need to be resolved during process-
ing. This gives one an impression that a medium PRF Pulse Doppler
radar due to its range 21:d doppler ambiguities would have a limited
use. In reality, it is not so With radar signal processing technology
making significant advances, it has become possible to resolve these
ambiguities. Along with pulse compression, it allows the designers to
build detection (requiring high energy, long transmitted pulses and
high PRF), Doppler (requiringhigh PRF) and Range Resolution (requir-
ing short processed pulses) capabilities. A typical medium PRF is in the
range of 10 to 20 kHz. An X-band radar operating at say 15 kHz would
have a maximum unambiguous range of about 10 k m and a maximum
unambiguous closing speed of 0.3 mach.
A pulse doppler radar employs n coherent radar system architec-
ture where the transmit signal and the receiver local oscillator are
synchronised to a highly stable reference. Coherent, Coherent-on-
receive and Non-coherent types of radar architectures have been dis-
cussed in detail in the earlier part of the chapter.
RADAR RJNDAMENTALS 457

9.16. True and Apparent Ranges

The Appareut Range which is the range corresponding to the time
difference between the received echo pulse and the last transmitted
pulse, is different from the True Rcnge if the target is not a zone-1
target. Fig. 9.59(a)shows the true and apparent propagation times for
a zone-2 target while Fig. 9.59(b) shows the same for a zone-4 target.
The two propagation times are interrelated by :
TA = T p MOD PRI
where PRI = Pulse Repetition Interval
TA = Apparent propagation time
T p = True propagation time

k R*--I R -I(b)

Fig. 9.59
The apparent and true target ranges are interrelated by :
RA = R MOD [c. PRI/2]
The above equations allow us to find apparent propagation time
or the apparent range from known magnitude of true propagation time
and range respectively. In practice, it will be required to find the true
values from the apparent values. The relevant expressions are :

The targets interpreted as being a t incorrect ranges due to am-

biguity are called ~ a r G eGhosts and the process of determining true
range from apparent range is called Range Deghosting. The equations
above also reveal that for targets that outside range zone-1 the apparent
range is a function of radar PRF. This phenomenon is used as the basis
for range deghosting in most radars. Thus, radars use multiple PRFs
to determine true range of ambiguous targets. The maximum range
which a given PRF can determine unambiguously corresponds to a PRF
value given by the highest common factor (H.C.F)ofdifferent PRFs used
by radar.
Occurrence ofBlind Range is yet another form of range ambiguity
when the radar fails to detect the target though the target is very much
there and is sending back to the radar a fairly strong echo signal. A
458 MICROWAVES AND RADAR
blind range occurs when the time of occurrence of a received echo pulse
coincides with the time of occurrence of a transmit pulse. Blind ranges
can be computed from :
RB = NR.c.PRI/~= NR.c/~.PRI
where 'NR' is an integer greater than zero. It equals 'n' for n-zone tar-
pet.
Problem 9.28 :A target is at a range of 100 k m . Determine the
apparent range as would be measured by a radar operating at a PRR of
10 kHz and not using any range deghosting measures.
Solution : PRR = 10000 PPS
Therefore, PRI = 1/10000 = 0.0001s
The apparent range 'RA' is given by :
RA = R MOD (c. PRI/2) where 'R'is the true range

Problem 9.29 : A radar operating at 2000 PPS and looking at a

range zone-3 target measures the range as 25 km. The radar does not
employ any range deghosting measures. Determine the true range of the
target.
Solution : Apparent range, RA = 25 km
PRF = 2000 PPS
The true range 'Ris given by :
R = RA+ ( d 2 ) [(NR- l)/PRFI
= 25 + (3 x lo5/2) [(3- 1)/2000] ... NR= 3 for a zone-3 target

Problem 9.30 :A radar observes a target at 100 km, 50 km and

20 k m respectively when operatingat 750PPS, 1000 PPS and 1250 PPS.
Estimate the true range.
Solution :For PRF = 750 PPS, RA = 100 km
The true range in this case for a zone-1 target would be given by :

The true range for this target if it was in zone-2, zone-3, zone-4,
zoned, zone-6 etc. can be computed from the above expression by
substituting 'NR' equal to 2, 3, 4, 5, 6 etc. The possible values of true
ranges are 100 km, 300 km,500 km,700 km, 900 km and so on. .

For PRF = 1000 PPS, RA= 50 km

The true range in this case for a zone-1 target would be given by :
RADAR FUNDAMENTALS 459

The true range for this target for the zones 2, 3,4,5, 6 etc. would
be 200 krn, 350 km, 500 km, 650 km, 800 km and so on.
For PRF = 1250 PPS , RA= 20 km
The true range for a zone-1 target in this case would be given by :
~=20+(3~10~/2)~0=20km
The true range for this target for zones, 2, 3,4, 5, 6 etc. would be
20 km, 140 krn,260 km, 380 km, 500 km, 620 km and so on.
Thus, the possible true ranges for radar measurements made at
750 PPS, 1000 PPS and 1250 PPS are given as under :
750 PPS 100 km, 300 km, 500 km, 700 km, 900 km, 1100 km
100 PPA 200 km, 350 km, 500 km, 650 km, 800, km, 950 km
1250 PPS 20 km, 140 krn,260 km, 380 km, 500 km, 620 km
The shortest possible range that has been measured at all PRFs
is 500 km. It can be vertified that the next coincident range would be
1100km. Now, the maximum possible range that can be unambiguously
measured using the given PRF combination is the one that corresponds
to a PRF value equal to H.C.F. of 750,lOOO and 1250 PPS. The H.C.F.
is 250 PPS and the range is 600 km. Therefore, the true range is 500
km and not 1100 km.
9.17. Tracking Radar
The primary function of a tracking radar, as the nama suggests,
is the automatic tracking of moving targets. It is usually a ground based
system used to track the airborne targets. The tracking radar antenna
sends out a very narrow beam whose width could be anywhere between
fraction of a degree to a degree or so in both azimuth and elevation to
get the desired resolution for tracking purpose. One can however
visualise that it would be necessary to acquire the target with a search
radar having a beam ofrelatively much large width before a track action
is initiated. In the track mode, whenever the target tends to move away
from the radar beam axis, an error signal is generated which in the
closed loop is used to steer the radar antenna either mechanically or
electronically to keep the target always illuminated by the radar beam.
Track Modes
Tracking could be carried out using Range (called Range Track-
ing), Doppler (Called Doppler or Velocity Tracking) and Angular (called
Angle Tracking)information. This allows the radar to follow the motion
of a target in azimuth and elevation (due to angle tracking), range (due
to range tracking) and doppler (due to doppler tracking). However, not
all radars track in all dimensions. Different track modes include:

1. Single Target Track (STT)

2. Spotlight Track
3. Multi-Target Track
4. Track-While-Scan
460 MICROWAVES AND RADAR

In the Single Target Track, the radar tracks a single target. It is

continuously dedicated to a single moving target. Such a radar samples
the target information at the radar PRF. Single target trackers are
capable of tracking targets with great accuracy. In the Spotlight Track,
the radar sequentially dwells upon various targets spending a certain
specified time on each target. It is not as accurate as the single target
track due to the fact that a given target is likely to undergo a change in
its co-ordinates during the time between two successive dwell periods.
A Multi-Target Track mode is capable of simultaneously tracking
multiple targets with an accuracy matching that of a single target track.
In a Track-While-Scan (TWS) system, the radar samples the position
of several targets once per scan and then with the help of certain
extrapolation algorithms, estimates the position of targets between
samples. TWS is truly a search radar's operational mode. It is not
essentially a tracking operation because for true multi-target tracking,
each target must be sampled at the Nyquist rates corresponding to the
radar servo loop and target maneuvering bandwidths. The required
sampling rate may typically be 10 to 20 samples per second for each
target. On the other hand, in a TWS process, the target may be sampled
once every 10 to 15 seconds.
Fig. 9.60 shows the basic block schematic arrangement of a track-
ing radar. Most tracking radars use angular information as the basis
for tracking operation. But for accurate tracking, it is important

* Receiver -
Angle Error
Demod.

Freq. & time Range/Dopp-

stan- er Gate Gen.

I Servo Control

Fig. 9.60
that the radar concentrates on one target at a time. If there are more
than one targets in the radar antenna's beam, techniques should be
used to ignore other returns from other targets. Range Gating (which
is a part of Range Tracker) and Doppler Gating (which is a part of
Doppler Tracker) can be used for the purpose. Range and Doppler
trackers are described in detail in the latter part of the chapter. In fact,
time and frequency control for range and doppler gatingis done in range
and doppler trackers respectively. The angular error signals for the
desired target to be tracked are developed in the error demodulator
block which is also controlled by Range / Doppler gate generation block
and then fed back to the steerable antenna in a closed loop for tracking.
RADAR FUNDAMENTALS 46 1

Tracking Radar Types

Tracking radars are classified based on the n~ethodologyused to
develop angular errors. The commonly used tracking methodologies
are :
1. Lobe Switching
2. Sequential Lobing
3. Conical Scan
4. Amplitude Comparison Monopulse
5. Phase Comparison Monopulse

Lobe Switching :In the Lobe Switching tracking technique, the

antenna beam is rapidly switched between two position around the
antenna axis in a single plane as shown in Fig. 9.61. The amplitudes of
the echoes from the target to be tracked for the two lobe positions are
compared. The difference between the two amplitudes indicates the
location of the target with reference to the antenna axis. When the
target is on the axis, the difference is zero as the echo amplitudes for
the two lobe positions are identical. To summarise, the amplitude and
sense of the difference signal can be used to generate the corrective
signal which with the help of servo loop control can be used to move the
antenna so as to bring the target on the antenna axis. The lobe switching
technique has the disadvantage that it loses its effectiveness if the
target cross-section changes between different returns in one scan.

Fig. 9.61
462 MICROWAVES AND RADAR

Sequential Lobing : In Sequential Lobing, a squinted radar

beam (a radar beam whose axis has been shifted slightly off the antenna
axis) is sequentially placed in discrete angular positions, usually four,
around the antenna axis (Fig. 9.62). The angular information on the
target is determined by processing several target echoes. Tne track
error information is contained in the target signal amplitude variations.
The squinting and squinted beam switching between different positions
is done electronically in modern radars using this tracking methodol-
ogy. Since the beams can be switched very rapidly using electronic
means, the transmitted beam is usually not scanned. The lobing is on
receive only. Also, virtually any scanning pattern can be used. The scan
pattern can be changed from scan to scan. It is because of this reason
that this type of tracking radar is less affected by amplitude modulated
jamming.

Fig. 9.62
Conical Scanning: This is similar to Sequential lobing described
above except for the differencethat in case of Conical Scan, the squinted
beam is scanned rapidly and continuously in a circular path around the
axis (Fig. 9.63). If the target to be tracked is off the antenna axis, the
amplitude of the target echo signal varies with the antenna's scan
position. The tracking system senses these amplitude variations as a
function of scan position to determine target's angular co-ordinates.The
error information is then used to steer the antenna axis so as to coincide
with the target location. For true tracking, the scan frequency must be
such that Nyquist criterion for the sampling rate is met. In a pulsed
radar, there must at least be four pulses per scan, two for generating
azimuth error signal and two for generating elevation error signal. This
implies that the maximum scan frequency can be one fourth of the radar
PRF. The actual scan frequency also depends upon the scan mechanism

Angle

Fig. 9.63
RADAR FUNDAMENTALS 463

- IC
PSD

Trons- 90
rnitter
I
Scan Scon Ref
Generator

Fig.9.64
and is smaller of the two values set by maximum scan rate dictated by
Nyquist criterion and the scan mechanism capability. The antenna
beam is squinted and scanned either mechanically by offsetting the feed
androtatingit or electronicallywith the help ofphase shifters. Mechani-
cal scans are usually much slower than electronic scans. Typical scan
rates are 30 to 40 scans per second.
Fig. 9.64 shows the block schematic arrangement of conical scan
system. The scan reference is a sinusoidal signal varying at the scan
rate and phase locked t o the scan signal. The functions of the other
building blocks are self explanatory.
Conical scan technique of tracking is highly vulnerable to
amplitude modulated jamming particularly the gain inversion jam-
ming. In gain inversion jamming, the target carrying the jammer
receives the radar's transmitted signal. The jammer demodulates the
amplitudevariation,inverts it and then sends it back towards the radar.
The radar starts tracking the jammer signal if the jammer signal is
much stronger than the echo signal. Since the jammer transmits
inverted angle information, the radar would never be able to track the
target.
Amplitude Comparison Monopulse Tracking : One of the
major disadvantages of sequential techniques including Lobe Switch-
ing, Sequential Lobing and Conical Scan is that their tracking accuracy
gets severely affected if the target's radar cross-section changes during
the time when the beam is being switched or scanned, as the case may
be, to get the desired number pf samples. In addition, these techniques
also suffer from their vulnerability t a AM jamming. Monopulse Track-
ing, both Amplitude Comparison and Phase Comparison, overcomes
these shortcomings by generating all the required angle error informa-
tion from one pulse only.
The basic principle of operation of Amplitude Comparison
Monopulse Tracking can be explained with the help of Fig. 9.65. Fig.
9.65(a)shows a radar antenna and the received wavefront when the
target is on the antenna axis. In this case, the received wave front will
be focused onto a spot on antenna axis as shown. If the antenna used
464 MICROWAVES AND RADAR

four feeds placed symmetrically around the focal point representing

four quadrants A, B, C and D as shown, then the amount of energy

Fig. 9.65
falling on each feed would be the same as shown in Fig. 9.66 (a).Now,
if the target is located off axis as shov;n in Fig. 9.65(b),then different
feeds would receive different energie: depending upon magnitude and
sense of this off-axis angle. For instance, if the target is above the axis
and having the same azimuthal locatim as the antenna axis, then the

Fig. 9.66
received energy would be distributed as shown in Fig. 9.66 ( b ) . Fig.
9.66(c) gives the received pattern when the target is below the antenna
axis but having the same azimuthal location as that of the antenna axis.
Figs. 9.66 ( d )and ( e )depict the condition when the target is to the right
and left of the antenna axis respectively with the target having same
elevation location as the antenna axis in both cases. Fig. 9 . 6 6 0 shows
a condition where the target is above and to the right of the antenna
axis. The amplitude of the received echo pulse at the outputs of various
feeds can be appropriately processed to determine azimuth and eleva-
tion error sigiiis along with some other useful error signals. One such
RADAR E'UNDAMENTALS 465

processing method is shown in Fig. 9.67. The azimuth and elevation

error signals in this case are respectively are given by :

A o(A+D)-(B+C)
Azimuth Error

C
I I o(A+B)-(C+D)
I I ElevationError

D-"q (A I s u r n Channel

U(A-B)+(c-D)
Fig. 9.67
Azimuth error signal = (A + D)- (B+ C)
Elevation error = (A + B) - (C+ D)
(A + B + C + D) gives the sum channel.
It is not necessary to pairA, D and B, C only as shown in Fig. 9.67.
One could pair A, B and C,D also as shown in Fig. 9.68. In that case,
the azimuth error would be difference of first differences rather than
difference of first sums and the elevation error would be difference of
first sums rather than sum of the first differences.

I, A d*+D)-(B+C)
Azimuth Error

I
D t Elevation Error

Fig. 9.68
In the amplitude comparison monopulse tracking technique, it is
important that signals arriving at various feeds are in phase. This is
not a problem when using reflector antennas with feeds that are
physically small, usually a few wavelengths across. In case of arrays
where the antenna surface is very large, signals arriving from different
off-axis angles present different phases to different segmentsinto which
the array has been divided. These phases need to be equalised before
error signals are developed.
Amplitude comparison monopulse is somewhat more complex to
implement with array antennas. Now for tracking, one needs to trans-
466 MICROWAVES AND RADAR

mit the transmit beam and four squintec' beams for error detection
which could be conveniently implemented by using a foui-horn or a
five-horn reflector feed. In case of an array, the array is divided into
four quadrants. Signals from each of the four quadrants feeds a beam
formingnetworkcomprisingof phase shifters qecessary to produce that
very squinted beam. These four beams are then fed to monopulse
comparator to generate sum, azimuth and eievation error signals
required for tracking.
Phase Comparison Monopulse Tracking : In Phase Com-
parison Monopulse Tracking, it is the phase difference between the
received signals in different antenna elements that contains informa-
tion on angle errors. In all, at least two antenna elements are required
each for azimuth and elevation error detection. When the target is on
axis (Fig.9.69a), the magnitude of phase difference would be zero. If it
is off-axis,then magnitude and sense of the phase difference would
determine the magnitude and sense of the off-axis angle (Fig. 9.693).
The sensitivity of this technique i.e. the phase difference produced per
unit angular error, increases if the elements are wide apart. But if they
are too far apart, an off-axis signal may produce identical phases at the
antenna elements (Fig. 9.70). This gives rise to ambiguity. A practical
system could have two pairs of antenna elements each for azimuth and

AqfT$
elevation. The outer pair gives the desired sensitivity while the inner
pair resolves ambiguity.

E~~~~ Phase Detector

(b)
Fig. 9.69
9.18. Range Tracking
Range Tracking is the process of tracking a moving target based
on its range co-ordinates. Even though the commonly used tracking
methodology in tracking radars is angle tracking, a range tracker forms
RADAR FUNDAMENTALS 467

a part of the angle tracker also. A range tracker in that case continuous-
ly measures the target range and based on the range data generates a
range (or time) gate so that the target is at the center of the gate. Range
tracking thus provides an effective means of distinguishing the desired
target to be tracked by using, may be, angular means from other targets
within radar beam. Doppler tracking discussed in the next section is
another.
The first step in any tracker, range or otherwise, is target acquisi-
tion which provides an idea about the target co-ordinates so that radar
beam could be pointed in that direction. Range tracker could do the job
of target acquisition very well. Typically, the range tracker divides the
minimum to maximum range into small range increments and as the
antenna scans a given region, it examines each of the range increments

Fig. 9.70
in a given direction simultaneously for presence of target. The antenna
is made to scan slowly enough for the target to remain within the radar
beamwidth as different range increments are being examinedin a given
direction.
Range tracker is a closed loop system. The error correspondingto
deviation of target's range location from thz center of the range gate is
sensed and fed back to the range gate generating circuitry to reposition
the gate in such a way that the target is at the center. The commonly
used technique of sensing range tracking error is that of using split gate
comprising of an 'Early Gate' and a 'Late Gate' as shown in Fig. 9.71.
When the target is at the center, the area under the echo pulse when
early gate is open is same as the area under the pulse when the late
gate is open. If the signal under the two gates are integrated and a
differencetaken, it would be zero. If the target is off-center, one signal
would be greater than the other. The magnitude and sense of the
difference signal can be used to reposition the gate.
468 MICROWAVES AND RADAR

Area under Area under

Early gate Late gate

i Range track Gate

I
I !
Early Gate

I
Time (Range)
Fig. 9.71
-
9.19. Velocity Tracking
Velocity Tracking is process that makes use of doppler shift infor-
mation. Doppler tracking error using a split filter error detection. The
track error is represented by the difference between target IF and
receiver's normal IF. The error after filtering is used to change the
receiver local oscillatorfrequency until doppler shifted signal is nominal
IF. Fig. 9.72 shows the block schematic arrangement in a velocity
tracker.

Trans- Waveform
I mitter Generator

Mixer - - IF
I

Split Filter
Error Gen.

Fig. 9.72
9.20. Pulse Compression Radar
It is always desirable to have long transmitted pulse in a radar to
enhance its detection capability. Along transmit pulse, for a given PRF,
increases the average power and reduces the ratio of peak power to
average power. In other words, for a given average power requirement
depending upon the required ranging capability, the peak power stress
would be less in case of radar asked to transmit longer pulse. On the
contrary, the range resolution capability of the radar is inversely
proportional to pulse width. A narrower processed pulse is desirable for
achieving higher range resolution. In the conventional pulsed radar
RADAR FUNDAMENTALS 469

where the transmit signal is gated CW pulse of a fixed frequency, the

transmit and received pulse widths are approximately the same and
therefore a higher detection capability could be achieved only at the
expense of poorer range resolution. We could have the best of both
worlds if somehow we could transmit a long pulse and have a narrow
processed pulse. This is exactly what is done in a Pulse Compression
Radar. A pulse compression radar transmits a wide pulse called Ex-
panded Pulse to achieve higher average power and detection capability
and processes the received pulse to a narrow pulse called the Com-
pressed Pulse. The compressed pulse width is approximately equal to
reciprocal of the signal bandwidth in the expanded pulse.
There are two basic techniques commonly used for pulse compres-
sion. One is analog in nature and the other is digital. In analog pulse
compression, the expanded transmit pulse, instead of being a single
frequency burst, has a frequency modulation across the pulse. It could
be an up-chirp where the frequency linearly varies from a low value to
a high value (Fig. 9 . 7 3 ~or
) a down-chirp where it varies from an initial
high value to a final low value (Fig. 9.73b). In the digital scheme, some

Fig. 9.73
kind of phase coding is given t o the R F across the pulse. The received
echoes are compressed either by using a matched filter concept where
the compression filter in the receiver is matched to the transmit wave
or by the process of correlation where a delayed version of transmit
signal is correlated with the received echo. Analog pulse compression
is done using matched filter concept while correlation technique adapts
better to digital pulse compression.
Fig. 9.74. shows the basic block schematic arrangement in case of
a pulse compression radar using matched filter concept. The diagram
is self explanatory except for a couple of points. The compression filter
response is complex conjugate of that of expression filter. From the
property that If H(w) is Fourier transform of h(t), then H* (a) is the
Fourier transform o f h (- t ) ,any phase change introduced into the signal
by H(a), such as linear FM, is undone in H* (a). So, if expansion filter
is fed with a short pulse with RF a t receiver IF (COHO), its output is
the expanded pulse with modulated RF. The received echo at the input
of H* (w) is again an expanded pulse with RF equal to doppler shifted
IF whereas its output is a compressed pulse whose width is ap-
proximately equal to the short pulse at the input of the expansion filter.
The RF of the compressed pulse equals the doppler shifted IF.
,-

470 MICROWAVES AND RADAR

Trans-
mitter

Fig. 9.74
Tllc compressed wave in practice is not exactly the same as the
COHO pulse that was expanded. Compressed pulse infact leaks into
time re;Tion other than that occupied by echo (Fig. 9.75~).This leads to
lev amplitude desired echoes getting hidden by large interfering sig-
n&. This necessitates introduction of mismatch (silmilar to a window)
to redme leakage (Fig. 9.75b).Rut in the process, the compressed pulse
+dth increases a bit. This can be compensated for by starting with a
pdse a bit shorter than the desired compressed pulse width.

Fig. 9.76 shows th" schematic arrangement for achieving pulse

compression in a radar employing digital pulse compression. A delayed
copy of transmit wave is being correlated with received echo to compress
the pulse. In principle, it is equivalent to matched filter concept.

T
-O
-, tranmitter

Circuit

compressed O/P
Corrclator

Fig. 9.76
RADAR FUNDAMENTALS 47 1

Fig.9.77
To achieve analog pulse expansion and compression,the commonly
used device is the SAW device. By proper design of interdigital
transducers, desired expansion filter characteristics can be achieved.
The compressed pulse is the matched version of it. Fig. 9.77 (a)shows
a typical SAW based expansion filter producing an up-chirp expanded
output. Matched compression filter is shown in Fig. 9.77(b). Detailed
discussion of SAW devices is beyond the scope of this text. In case of
digital pulse compression, a variety of codes are available. Barker code
bit sequences and Pseudo random bit sequences are the commonly used
ones. Details are again beyond the scope of the text.
9.21. Synthetic Aperture Radar
In an earlier part of the chapter, it was mentioned how the physical
size of a real antenna (physically realisable antenna) limits the mini-
mum achievable beamwidth. The minimum achievable beamwidth may
not be adequately small to yield cross-range resolution that would bc
desirable for some specific applications such as terrain mapping, imag.
ing etc. Synthetic Aperture Radar (SAR) synthesises a large .antenna
by moving a real antenna through different locations across the volume
of interest. The synthesised length then equals the distance moved by
the real antenna (Fig. 9.78). The antenna moves through different
locations and a pulse transmitted from each location. The received
echoes from different locations are stored and then processed in such a
way that the antenna behaves as if it was as long as the distance
travelled. If there was an antenna physically as long as the synthesised
antenna, the incremental elements of the synthesised antenna would
be formed simultaneously here and the narrow beam would be formed
by summing outputs from all those elements. In case of synthesised
antenna, these incremental elements would be formed sequentially.
The outputs from these elements are stored till the full array is formed.
The outputs are then summed to form the narrow beam. The two
situations are depicted in Figs. 9.79(a) and (b). '
472 MICROWAVES AND RADAR

Position- 3

L = Length of real antenna

position-l Leff= Effective length of synthetic
antenna

Fig. 9.78

Toraet Tor aet

I I
From From
tr onsmitter transmitter
Fig. 9.79
Cross-Range resolution of a synthetic aperture radar is given by :
Cross-Range resolution, & = [A.R / 2 . Leff]
Also, Leff = R h/Vef
Therefore,Xs= [A. R / 2 ] . [D&R . h ] = Veff/2
Lef = Effective length of synthesised antenna
D,r = Effective length of real antenna
The above equation for the cross range resolution implies the
following :
1. The cross range resolution is independent of range.
2. Cross range resolution is not a function of operating wave-
length.
3. A smaller rep1 antenna makes a better synthetic antenna.
RADAR FUNDAMENTALS 473

First statement is true because effective length of synthetic anten-

na increases with range. So, any degradation in cross range resolution
with range would get compensated by increase in 'Leff'. If synthetic
antenna has a larger beamwidth due to increase in operating wave-
length, the real antenna would also have increased beamwidth which
in turn means increased 'Leff*. Smaller real antenna means large
beamwidth and a large 'Leff '. Synthetic aperture radars are used for
high resolution ground mapping from airborne platforms.
Inverse Synthetic Aperture Radar
Inverse Synthetic Aperture Radar (ISAR) is an extension of the
concept of synthetic aperture radar. In ISAR, the radar is usually
stationary (it could be a moving one too) and it is the target motion, or
rotation to be more precise, that is made use of to synthesise a large
antenna. It takes advantage of the differential doppler shift that exists
between the echoes from different objects in the same range bin when
the target is rotaKng to enhance the cross range resolution. These are
commonly used for imaging tactical targets from airborne, shipboard
and land based platforms.

Problem 9.31 :A pulse compression radar transmits a linearly

frequency modulatedpulse that is 4 p wide. The RF frequency is swept
from 50 MHz to 70 MHz across the pulse. Determine the compression
ratio and the width of the compressed pulse.
Solution :Expanded pulse width, TE = 4 ps
Signal bandwidth, B = 70 - 50 MHz = 20 MHz
Compressed pulse width, zc = 1 / B = 1/20 = 0.05 ps
Compression ratio = .~E/TC
= 4/0.05 = 80

The compression ratio can also be computed from the given values
of expanded pulse width and the signal bandwidth. It is the product of
the two.

Problem 9.32 : A synthetic aperture radar operating at 10 GHz

uses an antenna having a n aperture of 2m. Determine the synthesised
aperture and the cross range resolution for a target at a rcmge of 10 km.
Solution : Operating frequency = 10 GHz
Therefore, operating wavelength, h = 3 x 108/10 x 10' = 0.03 m
Antenna aperture = 2 m
Therefore, 3 dB beamwidth = 0.03 / 2 = 0.015 radian
Range, R = 10 km = 10000 m
Therefore, Lefj= 0.015 x 10000 = 150 m
Cross range resolution, Xi = [R. h/2. Led
474 MICROWAVES AND RADAR

= 10000 x O.O3/2 x 150 = 1m

Cross range resolution can also be computed from the expres-
sion :
Xs = D q f 2

9.22. Over-The-Horizon Radar (OTHR)

The maximum ranging capability of a microwave radar gets
limited by the radar horizon even if there is adequate transmitted power
to receive a detectable signal. The shortcoming could be overcome to
some extent by elevating the height of the transmitting antenna as
shown in Fig. 9.80. But this is certainly not the solution when we are
talking about ranges of the order of thousands of kilometers. Over-The
Horizon-Radar (OTHR), as the name suggests, can look at the targets
that lie beyond the radar horizon. This is made possible by making use
of propagation modes other line-of-sight propagation used by
microwave radar. These include ground wave propagation and sky wave
propagation. The radar is operated in the high frequency (HF)band of
the electromagnetic spectrum. The frequency of operation therefoi e lies
between 3 and 30 MHz.

Fig. 9.80

Two types of OTHR have been developed known as the Surface

Wave (or Ground Wave) Radar and the Sky Wave Radar making use of
the surface wave and ionospheric propagation modes respectively.
Surface wave OTHR uses the surface propagation mode to look over the
immediate horizon. It is predominantly monostatic in nature. A mono-
static radar is the one that has a co-located transmitter and receiver.
Typically, such an OTHR has a maximum ranging capability of 400 k m
or so.
Skywave OTHR makes use of ionosphere to transmit electromag-
netic waves to very long distances beyond the radar horizon. It is
bistsatic in nature. A bistatic radar has its transmitter and receiver
widely separatedfiom each other. The radar beam is beamed up towards
the sky. It gets reflected from the ionosphere to illuminate the target.
The radar signal as reflected from the target reaches the radar receiver
RADAR FUNDAMENTALS 475

again via the ionosphere (Fig. 9.81). The radar transmitter and receiver
in an OTHR are typically separated by about 100 kms. As is clear,
OTHR can look at targets close to the radar site. Typically it can detect
targets between 1000 kms and 4000 kms. This type of OTHR is
particularly suited to the defence and remote ocean sensing needs of
Ionosphere

%t

Fig. 9.81

countries having very extensive geographical areas such as U.S.A.,

Russia, Australia etc. OTHR performance, due to the fact that it makes
use of ionospheric propagation, suffers from sun spot activity,
Daymight cycle, seasonal variations and so on.
Targets of interest for an OTHR are the same as they are for the
microwave radar which include aircraft, missiles, ships, other strategic
locations and so on. In addition, due to the fact that the wavelengths
used are of the same order as that of the ocean gravity waves, this type
of radar has also been used with great advantage in providing informa-
tion on the wave height directional spectrum and consequently by
inference on surface winds and ocean currents. It can also be effectively
used for observhgvarious forms of high altitude atmospheric ionisation
phenomenon such as that due to meteors, aurora, missile launches etc.
Fig. 9.82 shows the transmit (Fig. 9.82 a)and receive (Fig. 9.82 b ) sides
of an over-the-horizon-radar (TypeANIFPS-118).The systemis capable
of providing electronic surveillance of aircrafts at extended ranges of
800 km to 2880 km.Operating in the HF band between 5 MHz and 28
MHz where the radar energy is reflected by the ionosphere, this system
is designed for over-the-horizondetection and 'trackingof aircrafts and
cruise missiles flying at any altitude. The transmit and receive sites are
separated by about 150 km to 200 km.
 ND RADAR

3.04

9.23. Monostatic and Bistatic Radars

In the basic Monostatic Radar system, the radar transmitter and
receiver are co-located and usually a single antenna is used for both
transmission and reception. A duplexer is used to separate the transmit
and receive functions both in time and power amplitude domains. Fig.
9.83 shows the basic monostatic radar system.
RADAR FUNDAMENTALS 477
Antenna

Duplexer

Transmitter

Fig. 9.83
In the Bistatic Radar system, separate antennas are used for
transmission and reception purpose and the two antennas may some-
times be hundreds of
kilometers apart as
is the case in over-
the-horizon radar.
Fig. 9.84 shows the
basic set-up. In a
typical set-up, the
receiver receives
both directly from
the transmitter as
well as from the tar- Fig. 9.84
get. The two
iropagation times as well as azimuth and elevation angle measure-
ments at the receiving site give the target location.

I I
I I
II
*To Data I
Processor & I
Display

lator

9.24. Primary and Secondary Surveillance Radars

A Primary Radar is the one that we have been discussing until
now. It makes use of reflection from the target to determine various
target related parameters and characteristics. A Secondary Surveil-
lance Radar differs from a primary radar in the sense that it does not
make use of reflection of the transmitted energy. In a secondary
478 MICROWAVES AND RADAR

surveillance radar system, pulse transmission from ground is received

at the target, usually an aircraft, detected and decoded in the
transponder. The aircraft transponder then transmits coded pulses
back to ground after a certain known delay. Thus identity of the aircraft
can be established on ground. A r a d a ~of this kind serves an extremely
useful purpose in identifying friendly aad enemy aircraft. IFF system
(Identificationfriend or Foe)is another name for such systems. Fig. 9.85
shows typical block schematic arra2gement of an SSR system alone
with the typical block representation of aircraft transponder.
Performance Comparison Of Primary and Secon-
dary Surveillance Systems
1. In terms of positional data, primary radar is characterised by
comparatively better resolution whereas SSR system has a
higher range accuracy potential.
2. Identificationoftarget capabilityis inherent in an SSR system
whereas the same has to be irierred from the position data in
case of primary radar system.
3. In a primary radar system, the, received signal is a function of
equivalent echoing area and thus the target size and shape.
In an SSR system, the signals from even small aircraft are as
strong from large aircraft.
4. SSR system is completely free from the constraints of ground
clutter as the ground station is not meant to receive the
transmitted frequency and reflected energy if any is totally
rejected. On the other hand, primary radar has to include
clutter rejection system. This can sometimes inhibit target
detection in high clutter regions.

Fig. 9.86
RADAR FUNDAMENTALS 479

5. An SSR system has a very high degree of dependence on the

target co-operation and is completely useless if the target does
not co-operate. That is, targets must have properly working
transponders for an SSR system to work. A primary radar
functions as long as there is a target present within its detec-
tion range.
6. Range capability of an SSR system is the compound of two
virtually independent inverse square law functions, one for
the interrogation range and the other for the reply range.
Primary radar range, when a single antenna is used, is an
inverse fourth power law function.
Fig. 9.86 shows the photograph of an air route surveillance
radar. It is capable of providing both primary as well as
secondary surveillance functions.
9.25. Laser Radar
A Laser Radar uses a laser beam instead of microwaves. In other
words, the transmitted electromagnetic energy lies in the optical
spectrum in laser radars whereas in microwave radars, it is in the
microwave region. The frequencies involved therefore are very high
ranging from 30 THz to 300 THz and the corresponding wavelengths
ranging from 10 pm to 1 pm in practical systems. This results in
tremendous increase in angular resolution. Also, as the operating
frequency increases, so does the available absolute bandwidth and thus
the time or range resolution. It is feasible to generate laser pulses as
narrow as a picosecond. Another advantage of laser radars is their
immunity to jamming when compared to microwave counterparts.
Laser radars can be used for all those applications where a microwave
radar has been used. These include target detection and ranging, target
characterisation and imaging and so on. Details are beyond the scope
of this text.
9.26. Radar Guidance of Missiles
There are three broad categories of radar guidance of weapons,
missiles being a typical representative. These include the following :
1. Beam Rider Guidance
2. Command Guidance
3. Homing Guidance

Each one of them is briefly described below.

Beam Rider Guidance
There are two forms of beam rider guidance namely Line-Of-Sight
Beam Rider (LOSBR) and Command to Line-Of-Sight Beam Rider
(CLOSBR).
In the Line-Of-Sight Beam Rider (LOSBR), a tracking radar con-
tinuously tracks the target and it is imperative that the target remains
illuminated by the radar beam right from the time of launch till the
480 MICROWAVES AND RADAR

engagement' takes place. A rear looking receiver in the missile tracks

the position of the missile relative to radar beam and thebeam riding
missile with the help of a gyro carried in the missile navigates the1 line
of sight established by the radar beam with the help of associated servo
control. The missile thus rides the radar beam and finally hits the target
being illuminated by the radar beam. Fig. 9.87 depicts the LOSBR
technique. Several missiles can be made to ride the same beam if they
are aimed at the same target.

Tracking
Radar

Fig. 9.87
The Command to Line-Of-SightGuidance (CLOS)requires a tar-
get tracker and a missile tracker situated at the base or the aiming
point (Fig. 9.88). The zero axis of the missile tracker is laid parallel to
the zero axis of the target tracker. The zero axis of the target tracker i s
maintained to be on the target by the tracking process. Missile is laun-

Missile Target

Missile

Fig.9.88
ched towards the target and the with the help of a flare or beacon
emitted from the rear of the missile, missile tracker senses any displace-
ment of missile from line of sight. The corresponding signal in a coded
form is sent to the missile over a communication link. The signal is
decoded in the missile and a suitable lateral thrust applied to put the
missile back on line of sight. Both target and missile must be tracked
until impact occurs. Again based on the type of trackers used for target
and missile, CLOS may be classified as Manual Command to Line-Of-
Sight (MCLOS),SemiAutomatic Command to Line-Of-Sight (SACLOS)
or Automatic Command to Line-Of-Sight (ACLOS).
RADAR FUNDAMENTALS 481

Command Guidance
In Command Guidance of missiles depicted in Fig. 9.89, there are
two separate tracking radars as in the case of CLOS, one for tracking
the missile and the other tracking the target. The missile and target
could also be tracked by two separate beams of the same phase array
type tracking rad,ars. The missile like in case of beam rider guidance
does not perceive the target. The states of the missile and target are
continuously fed to a computer which processes this data and then
generates the trajectory for the missile so that it intercepts the target.
The computer sends guidance commands to the missile continuously till
it intercepts the target. The accuracy of a such a system is inversely
proportional to the range as a fixed angular error gives a larger linear
error at the target for increased range. This system differs from CLOS
system in the sense that target tracker and missile tracker are capable
of lookingin totally different directions. In such a system, the trajectory
of missile is not coniined to line of sight and the missile may be
controlled onto the most effective path to impact the target.

Guidance

MissileTrack
Radar Computer
H Target Track
Radar I
Fig.9.89
Homing Guidance
Unlike the beam rider and command guidance techniques where
the target tracking is done by some means outside the missile, in
Homing Guidance,the missile usually has its own radar (calledSeeker).
The missile radar illuminates the target and extracts the tracking data
from the target reilected energy. The homing guidance is much more
accurate as for a fixed angular error, the linear error keeps reducing as
the missile closes on to the target. The increase in accuracy is however
at the cost of increase in missile hardware complexity. ~ h & aree three
types of homing guidance namely.
1. Active Homing
2. Semi-Active Homing
3. Passive Homing
482 MICROWAVES AND RADAR
In the Active Homing Guidance, the nlisile has its own radar
which illuminates the target. It is a fully seif'contained system. The
missile homes on to the target reflected energy.
In the Semz-Active Homzng GuzLZartce (Fig. 9.90), the target is
illuminated by a radar located elsewhere. It could be ground based or
airborne. The target is illuminated throughout the engagement. The
missile receives the illumination directly through its rear placed anten-
na as well as illumination reflected from the target through its front
end antenna. The two signals are processedinsidethe missile to extract
the guidance information. The missile finally homes on to the target
reflected energy.

Fig.9.90
Passive Homing depends upon some form of natural energy
emitted by or reflectedfiomthe target. There are many examples where
passive homing is a relatively cheap and effective method of guidance.
Anti-radar missiles for instance may home on to transmission from the
radar itself.
RADAR FUNDAMENTALS 483

Review Questions
1. Briefly describe the principle of operation behind the use of a
radar for measuring target range, velocity and location.
2. What do you mean by the terms 'Range Ambiguity' and 'Dop-
pler Ambiguity'? How can these ambiguities be resolved ?
3. "In case of a conventional pulse radar, an attempt made to
enhance the ranging capability of the radar results in loss of
rangeresolution."Elaboratethis statement in detail andname
any radar type that does not suffer from this problem.
4. Differentiate between 'Range Resolution' and 'Cross Range
Resolution'. Outline the radar parameters that influence
range resolution and cross range resolution.
5. What is a Synthetic Aperture Radar (SARI ? What is its
principle of operation and what are its main applications ?
6. With the help of radar range equation, briefly describe the
inpact of Peak Transmitted Power, Antenna Gain, Target
cross-section and Receiver (S/N) on the target range detection
capability of the radar.
7. Differentiate between coherent, Coherent-on-Receiveand In-
coherent radar transmitters.
8. Name the CW radar type capable of measuring both target
range as well as target velocity. Briefly describe its principle
of operation.
9. Low PRF radars are unambiguous in range but ambiguous in
doppler whereas high PRF radars are unambiguous in doppler
but ambiguous in range. Comment !
10. What is a Pulse Compression Radar ? How does it achieve the
higher range detection capability of a long pulse radar and
higher range resolution of a short pulse radar ?
11. Write short notes on :
(a) Blind ranges and Blind speeds
( b ) Range granularity and Doppler granularity
(c) Range ambiguity and Doppler ambiguity
(d) A-Scan and PPI
12. Differentiate between
(a) Monostatic and Bistatic radars
( b ) Primary and Secondary Surveillance radars
(c) CW and FM-CW radars
(d) MTI and Pulse doppler radars
13. What are the major problems encounteredin case of sequential
lobing and conical scan type of tracking methodologies ? How
are they overcome in a Monopulse tracking radar ?
14. Briefly describe the following modes of radar guidance of
missiles
484 MICROWAVES AND RADAR

(a)Beam rider guidance (b) Command guidance

(c) Homing guidance
Problems
1. Determine the maximum unambiguous range (assuming a
zone-1 target) and range resolution of a pulse radar transmit-
ting 10 ps wide pulses at a PRF of 1000 Hz.
(150Km, 1.5 K m )
2. A pulse Doppler radar operating at 3 cm wavelength has a
PRF of 1000 Hz. Determine the maximum radial speed of the
target which this radar can unambiguously measure.
(7.5 d s )
3. Aradar is looking at two targets having the same radial range
of 10 km. If the emitted beam has a 3 dB beamwidth of 2",
determine the minimum separation of the ,wo targets in the
cross-range allowed for the targets to be resolved.
(349m)
4. Determine the first three blind speeds of an MTI operating at
10 GHz with a PRF of 1000 Hz.
(15mls, 30 mls, 45 d s )
5. A pulse compression radar transmits a 5 ps wide linear FM
pulse. The RF is swept from 50 MHz to 70 MHz across the
pulse. Determine the range resolution of this radar.
(7.5 m)
6. A certain pulse radar has a maximum ranging capability of
100 Km for targets having a radar cross-section of (A) m2. If
the peak transmitted power is increased 16 times, antenna
gain (assuming same antenna is used for transmission as well
as reception) is increased 4 times, determine the new m d -
mum range detection capability of the radar for targets whose
radar crass-section is ( ~ 1 4m2
) (100km)
7. A radar operating at a PRF of 1kHz looking at range zone-2
target measures the range as 30 km. Determine the true range
assuming that no range deghosting measures have been
adopted in the radar.
(180km)
8. A radar operating at 2 kHz looks at a zone-3 target having a
radial range of 175Km. Assuming that radar does not employ
any range deghosting measures., determine the target range
as measured by the radar. (25 km)

9. A radar is to have a maximum range of 250 Km. Determine

the maximum allowable pulse repetition frequency for unam-
biguous reception. (600Hz)
RADAR FUNDAMENTALS 486

10. A pulse radar measures the round trip propagation time with
an error of 1 ps. What will be the corresponding error in the
measured range ? (150 m)

SELF EVALUATION EXERCISE

I. Multichoice Questions
1. A pulse radar with a transmitted pulse width of 10 micro-
seconds would have a range resolution of
(a) 150 m ( b ) 1.5 km
( c ) 15.m (d)indeterminate from given data
2. A pulse compression radar is operating with expanded and
compressed pulse widths of 5 microseconds and 100 nano
seconds respectively. Its range resolution is
( a )15 m ( b )150 m
( c ) 1.5 km (d)indeterminate from given data
3. The expanded pulse in a pulse compression radar is 4
microseconds wide. The RF is swept from 50 MHz to 60 MHz
across the expanded pulse. The compression ratio is
( a )40 ( b )200 ( c ) 240 ( d )none of these
4 . The maximum PRF (fr) that can be used for a maximum
unambiguous range (Rm,) is given by
( a )fr = c / 2 . R,, (b c/Rmax
( c ) 2. c/Rnia ( d )Rm,/2. c
5 . A ship sailing at a radial speed of 10 m/s towards an HF radar
operating at 3 MHz would produce a doppler shift of
( a )0.1 Hz (b)0.2 Hz (c) 1 Hz ( d )2 kHz
6. One of the following is an HF radar system
( a )MTI ( b )Pulse Doppler
( c ) FM-CW ( d )Over-The-Horizon Radar
7. One of the followingradar systems has both a higher detection
capability as well as a higher range resolution.
( a )Pulse Compression Radar ( b )Moving Target Indicator
( c ) Pulse Doppler Radar ( d )TIacking Radar
8. When the target is moving with a speed equal to the lowest
blind speed, the doppler shift equals
( a )twice the PRF ( b )one half the PRF
486 MICROWAVES AND RADAR

(c)equals PRF (d)none of these

9. One of the following is surely a tracking radar
(a)Monostatic radar (b)Monopulse radar
(c)Bistatic radar (d)MTI
10. In a pulse radar, the peak transmitted power is increased by
a fi.dor of 81, keeping all other parameters unchanged. The
maximum range determining capability of the radar increases
by a factor of
( a )4 (b)3 ( c ) 81 (49
11. ln a doppler radar, the transmitted frequent$ was 100 MHz
and the received echo frequency was 110 MHz. This indicates
%hat
(a)target is moving away from the radar
(h) target is moving towards the radar
(c)target is moving towards the radar with a radial velocity
of 10 d s
(d)target is moving away from the radar with a radial
velocity of 10 d s
12. If the target cross-section is rapidly changing, the best choice
for accurate tracking is
(a)Monopulse tracking (b)Conical scan tracking
(c)Lobe switching (d)Sequential lobing
13. The guidance technique iikely to have the least linear error at
the target is
(a)Command guidance (b)Beam rider guidance
(c) Homing guidance (d)None of these
14. The radar that gives a much higher cross range resolution
than what is inclicated by the size of its antenna is
(a)Pulse doppler radar (b)Pulse compression radar
(c)Tracking radar (d)Synthetic aperture radar
15. Range resolution of a radar is indicated by which radar
waveform parameter ?
(a)Transmitted pulse width
(b)Transmitted RF
(c)Processed pulse width in the receiver
(d)Radar beamwidth
RADAR FUNDAMENTALS 487

16. A narrow radar beamwidth indicates

( a )higher range resolution
( b )higher cross-range resolution
( c )higher doppler resolution
( d )poor tracking accuracy
17. A synthetic aperture radar operating at 10 GHz uses an
antenna whose effective aperture is 5 m. Its cross-range
resolution would be
( a )2.5 m

( d )indeterminate from given data

18. A moving target produces a doppler shift of 1kHz in the radar
signal. What should be the minimum PRF of the radar so that
this doppler shift is measured unambiguously ?
( a )500 PPS (b) 1000 PPS
( c ) 2000 PPS ( d )4000 PPS
19. A radar in which the radar beam is steered electronically is
( a )Tracking radar
( b )MTI radar system
(c) Phase array radar
( d )Synthetic aperture radar
20. One of the possible solutions to blind speed problem is to
( a )use an MTI radar system
( b )use a low pulse repetition rate
(c) vary the pulse repetition rate

( d )use a high pulse repetition rate

11. Read the following statements carefully and
identify True and False statements
1. A Low PRF radar is likely to be highly ambiguous in doppler.
2. A radar with a high range resolution is bound to have a high
cross range resolution also
3. For a given target speed, a target moving at right angles to
the radar beam axis produces maximum doppler shift.
488 MICROWAVES AND RADAR
4. While radar's range resolution is determined by the width of
the radar beam in azimuth, the cross range resolution is
determined by the processed echo's pulse width.
5. A synthetic aperture radar is characterised by a very high
cross range resolution.
6. A high PRF radar such as pulse doppler radar is likely to be
highly ambiguous in range unless special measures are taken
to resolve range ambiguities.
7. PRF stagger is used to overcome the problem of blind speeds.
8. An unmodulated CW radar can not be used to detect target
range.
9. In the semiactive mode of homing guidance, it is imperative
that the target is kept illuminated by the radar till the time of
the impact.
10. A pulse compression radar achieves a high range resolution
capability at the expense of ranging capability.
 Satellites And Satellite
Communication
communication satellite in essence is a microwave link

A repeater station placed in some slot in space a d orbiting the

planet earth with an angular velocity depending upon the
orbital characteristics including orbital radius which is a func-
tion of the height of the satellite above the surface of the earth and the
shape of the orbit whether the orbit is circular or elliptical, equatorial
or inclined, symmetrical or eccentric and so on. We have both active as
well as passive satellites. A passive satellite is just a simple reflector
that reflects a small fraction of the electromagnetic energy beamed at
it from a ground based station back towards earth for the intended
receiving stations. Active satellites on the other hand use active
electronic means to amplify the received signal, do frequency transla-
tion before re-transmitting the signal back towards earth.
Satellite communication owes its origin to a British science fiction
writer Authur C. Clarke who in one of its papers published in Wireless
World in 1945 proposed that a satellite orbiting the earth with an
angular velocity exactlyequal to the angular velocity ofrotation of earth
about its axis could form the basis of future global communication
systems. He went on to state in that paper that such a satellite would
drive the electric power needed for various sub-systems on board the
satellite from solar radiation.
The technology of satellites and their application potential have
grown enormously after the launch of first satellite in the form of
Sputnik-I in 1957. Today, the word satellite sounds so familiar to all of
us irrespectiye of our educational and professional background. It is a
subject of interest and discussion not only to electronics and com-
munication engineers, it fascinates hobbyists and other electronics
enthusiasts too. The reason for this remarkable transformation lies in
the manifold increase in the application areas where the satellites have
been put to use. The horizon of satellite applications has extended far
beyond providing intercontinental communication services. Some of the
most significant and talked about applications of satellites are in the
fields of remote sensing, earth observation, weather forecasting and so
on. Satellites have vast potential in their role as the guide to the weather
man giving the weather forecast. In addition, they have a whole range
of applications related to atmospheric monitoring and space explora-
tion, besides doing spying activities for some of the developed countries
who can afford such services.
490 MICROWAVES AND RADAR

A satellite today is being accessed today not only by giant sized

antennae located at large installations but also by roof top antennae. A
satellite today not only provides long distance audio and video com-
munications, it is playing a significant role in the newer communication
services such as data communication, fax, cellular telephone etc. They
also assist our ships and aircrafts navigate safely in all weathers. Earth
observation and remote sensing satellites keep us informed about
weather, ocean conditions, sea ice destructive storms, tidal conditions,
volcanic eruptions, pollution monitoring, monitoring health of our crops
and forests. Another class of satellites keeps watch on military activity
around the globe and provides great help in enforcing or policing arms
limitation agreements. Although mankindis yet to travel beyond moon,
the satellites have crossed the solar system to investigate various
planets other than Pluto. These satellites for astrophysical applications
with giant telescope on board have sent data leading to many new
discoveries about our universe. The application areas are multiplying
and so are the quantum of applications in each one of these application
areas.
This chapter gives an overview of satellites and satellite applica-
tions. The chapter begins with an introduction to some of the important
fundamental concepts like satellite orbits and orbital requiremcnts
with reference to satellites to be launched for different application
missions, satellite stabilisation requirements and techniques, satellite
launch procedures and so on. This is followed by a brief introductionto
different segments of a satellite link touching the earth segment and
the space segment. The last part of the chapter is devoted to satellite
applications covering communication applications, remote sensing and
earth observation applications, weather forecasting, scientific and tech-
nological applications and military applications. Like other chapters,
important concepts have been further illustrated with the help of a large
number of solved ~roblems.
The important topics covered in this chapter include :
Satellite Frequen- 10.10. Corn onents of a
cy Bands ~ a t e h t e Com-
What is a Satellite munication
Satellite Orbits Systm
and Trajectories 10.11. Multiple Access
Different T p e of Systems
Satellite Or its 10.12. Satellites for
Putting a Satellite Communication
i n the correct Services
orbit 10.13. Satellite for
Different Satellite Weather Patrol
- Orbital Require- and Atmospheric
Monitoring
ments
Satellite Stabili- 10.14. Remote Sensing
sation Applications
Orbital Parame- 10.15. Scientific and
ters Technological Ap-
Eclipses plications
10.16. Military Applica-
tions
SATELLITES AND SATELLITE COMMUNICATION 491

10.1. Satellite Frequency Bands

The frequency bands commonly used for satellite con~munication
for commercial applications include the C-band (614 GHz) with an
up-link frequency of 5.925 GHz t o 6.425 GHz and the down-link fre-
quency of 3.7 GHz to 4.2 GHz, Ku-band (1412 GHz) with an up-link
frequency in the range of 14.0 GHz to 14.5 GHz and the down link
frequency in the range of 11.7 GHz to 12.2 GHz and the Ka-band (30/20
GHz) with the up-link frequency in the range of 27.5 GHz to 30.0 GrIz
and the down link frequency in the range of 17.7 GHz to 21.2 GHz. The
frequency bands for military use include the microwave band with an
uplink frequency of 292 MHz to 312 MHz and a down link frequency of
250 MHz to 270 MHz, X-band with an uplink frequency of 7.9 GHz to
8.4 GHz and a down link frequency of 7.25 GHz to 7.75 GHz and the
Ka-band with an uplink frequency of 43.5 to 45.5 GHz and a down link
frequency of 20.2 to 21.2 GHz.
10.2. What is a Satellite
A satellite, in general, is any natural or artificial body moving
around a celestial body. In IFe present context, we are referring to
artificial satellites orbiting around earth. These satellites are put into
an orbit and have a pay-load depending upon the intended application.
It could be used as a sort of repeater station while taking in
reference to communication satellites or as a photographer taking
pictures of regions of interest during its periodic motion, or even taking
photographs of clouds and monitoring other atmospheric parameters,
thus assisting the weatherman in making a timely and accurate
forecast. It could effectivelydo the job of a spy in some military satellites
meant for the purpose, or that of an explorer when launched for
astrophysical applications.
The idea of satellites originated from the proposal submitted by
Arthur C. Clarke, as outlined in the introduction, to the Council of
British Interplanetary Society in 1945 in which he talked about what
is known as a geostationary orbit and where he emphasised the impor-
tance of this orbit that had an orbital period equal to the time taken by
earth to complete one rotation about its axis. He highlighted the
importance of such artificial satellites carrying the required instrumen-
tation to provide intercontinental communication services. Though the
idea of satellites originated from the desire to put an object in space
that would stand still with respect to earth, thus making possible host
of communication services, there are many more varieties where the
satellites need not have to be stationary with respect to the earth t o
perform the intended function, and thus have much lower orbital
height.
10.3. Satellite Orbits and Trajectories
Before we go on to discuss satellite orbits and trajectories, we
would outline some of the commonly used terms with reference to orbits
492 MICROWAVES AND RADAR

and trajectories. A broader compilation of satellite and satellite launch

relevant terms and definitions is kontained in the glossary given
towards the end of the chapter.
While a trajectory is a path traced out by a moving body, an orbit
is a trajectory that is periodically repeated. An inter-planetary trajec-
tory is the resultant trajectory determined by the sun and the interac-
tion of the moving body with all other members of solar systems, i.e.
planets. A launch trajectory is the path taken for instance by a rocket
which has been launched from earth's surface. More precisely, the
trajectory of a moving body is the movement of body's centre of gravity.
The physical ooncepts involved in the launching of a satellite were
first outlined by Newton in the beginning of eighteenth century. Refer

Fig. 10.1
to Fig. 10.1. The inner circle represents the earth's surface, 'C'is the
centre of the earth.
If we try throwing an object from the top of a mountain in the
horizontal direction with a certain velocity, we find that the object
traverses a certain trajectory and lands on the surface of the earth at
some point, say 'D'. If we launch the same object with progressively
greater horizontal velocities, we find that it lands further and further
away.
SATELLITES AND SATELLITE COMMUNICATION 493

Going a step further, if we keep increasing the velocity, we shall

reach a stage where the object lands just diametrically opposite to the
point from where we launched it (point G ) . And if the velocity is
increased still further, then the object will complete a whole orbit
around the earth and come back to the starting point. It will then
continue indefinitely to move in this orbit without slowing down, thus,
creating an orbiting satellite.
What keeps the satellite continuously moving in the orbit around
the earth can be explained 60m the basic principle of physics which
states that when a body is spun round in a circle, there is a force acting
on it directed from the centre of the circle, the magnitude of force
depending upon the speed with which it is going round. If 'm' and 'v' are
respectively mass and velocity of the moving body and 'R'the radius of
the orbit, then the force acting outwards from the center known as the
centrifugal force, is given by
mu 2
-
R
This centrifugal force is equal to but opposite in direction to the
centripetal force that acts towards the center of the circle. It is the force
of gravity which acts as the centripetal force. This is the basis of motion
of satellites round the earth. The gravitational attraction between the
earth and the satellite is balanced by the centrifugal force acting
outwards on the orbiting satellite. This is demonstrated in Fig. 10.2.

Fig. 10.2
494 MICROWAVES AND RADAR

It can also be argued without going into detail that the horizontal
velocity required to touch down at a certain point on earth's surface is
also a function of the height of the launch point above the surface of the
earth. So, we can also say that the minimum horizontal velocity re-
qttired to make an object orbit arountl the earth is afunction ofits height
above the surface of earth. So, all t.iat a launcher does when it wants
to launch a satellite is to take the satoll~teto a high enough altitude, and
then launch it with the required horizontal velocity.
Table 10.1 gives the minimum horizontal velocity required to put
an object into a circular orbit. This velocity, called the first orbital
vc!ocity (vl), depends only w o n the height ( h )from where the object is
released. The same i~rrue for the orbital period (T).
Table 10.1 also shows that when the object is released from a
height of 35,786 kms from the earth's surface, the consequent orbit has
an orbital period of 1436 minutes, that is 23 hours 56 minutes, which
is the same as the orbital period of one sidereal day. A sidereal day is
earth's rotation period measured with reference to distant stars rather
than the sun.
For a certain horizontal velocity, called the second orbital velocity
( v z ) ,the object has sufficient velocity to escape the earth's gravitational
pull. Table 10.2 shows variation of v2 as a function of height (h).For all
horizontal velocities between v l and v2 the object remains a satellite of
the earth. Mathematically,
v2 = \I2 V l
The trajectory becomes parabolic for horizontal velocities greater
than v2. The first orbital velocity can also be computed from :
v, =m
where 'p' is a constant equal to product of gravitational constant and
mass of earth, and R is the radial distance of the orbit from centre of
earth.
Table 10.1

I Height fi) (krn) 1 VI m / s / T (rnin)

I
SATELLITES AND SATELLITE COMMUNICATION 495

Relevent to satellite orbits, another parameter that is of sig-

nificance is the orbit eccentricity e. For horizontal velocity equal to the
Table 10.2

first orbital velocity, the orbit is circular and eccentricity is zero. For
horizontal velocities greater than first orbital velocity and less than
second orbital velocity, the orbit is elliptical and orbit eccentricity is
between 0 and 1. For horizontal velocities closer to second orbital
velocity, e is closer to 1.For horizontal velocity equal to second orbital
velocity, orbit is parabolic and e = 1,and for horizontal velocity greater
than second orbital velocity, orbit is hyperbolic and e > 1.
10.4. Different Types of Satellite Orbits
A variety of orbits are used for satellites depending upon their
mission requirement. The orbits used for communication satellites
invariably belong to one of the undermentioned types.
(a)Geosynchronous orbit
( b )Geostationary orbit
(c) Molniya orbit
In a geosynchronous orbit, the satellite motion is synchronised
with the earth's rotation, that is, the angular velocity of the satellite
equals the earth's rotation rate with the result that i t appears to be
stationary to an observer on earth.
It may be mentioned here that a geosynchronous orbit is not
necessarily geostationary orbit though the two terms are frequently
used interchangeably. A geosynchronous orbit may be circular or ellip-
tical and may be inclined to the equatorial plane.
Ageostationary orbit is a unique case of a geosynchronous orbit in
the sense that this orbit is circular and the orbital plane lies in the plane
ofthe equator. It is this extra feature that makes it truly geostationary.
Also, the satellite has the same direction of rotation as the earth. It can
be verified that a satellite orbiting the earth at height of about 35786
kms from the surface of earth in a circular equatorial orbit in the same
direction as the earth coppletes its rotation around the earth once in
the same time as that taken by earth to rotate around its own axis once
(Fig. 10.3).
 Fig. 10.3.
The main advantages of a geostationary orbit which make it the
favourite of communication satellites include the following-:
(i) Since the satellite remains stationary with respect to an
observer on earth, there is no need to have costly tracking
equipment.
(ii) A relatively small number of satellites are needed to provide
continuous and uninterrupted service.
(iii) Due to a large coverage area, a large number of earth stations
can intercommunicate.
(iv) Doppler shift is nearly zero due to no relative motion between
the satellite and the earth station.
The disadvantages, though far outweighed by the advantages
outlined above, include :
(i) This orbit does not offer coverage of higher latitude regions.
Typically, latitudes greater than 80" north and south are not
covered.
(ii) The signal propagation time is very large, about 270 ms on an
average.
(iii) With increasing altitudes, the effects of eclipses become
pronounced.
(iv) The earth stations need to have highly sensitive and low noise
receivingfront ends as thereceived signal strengthis inversely
proportional to square of the distance between the earth
station and the satellite.
Another commonly employed satellite orbit is the sun synchronous
orbit where the plane of the orbit maintains a fixed orientation with
respect to the sun. The idea is that different parts of the globe would be
SATELLITES AND SATELLITE COMMUNICATION 497

traversed at more or less the same time of the day. Sun synchronous
orbit is popular with earth observation satellites.
Molniya orbit serves the purpose of a geosynchronous orbit for high
latitude regions. It is a highly eccentric orbit designed to spend some
eight hours above a particular high altitude station before diving down
to a low level perigee at an equally high southern latitude (Fig. 10.4).

/
S
Fig. 10.4
Usually three satellites at different phases of the same orbit are
capable of providing a continuous uninterrupted service. However,
controllable ground antennae are required because satellites do not
remain accurately on station, and also because it is necessary to switch
satellites before each plunges.
Based on the orientation of the orbital plane, the satellite orbits
are characterised as equatorial orbits where the orbital plane coincides
with the equatorial plane [Fig. 10.5(a)],polar orbits where the orbital
plane coincides with the polar plane [Fig. 10.5 ( b ) ]and inclined orbits
where the orbital plane makes a known angle with the equatorial plane.
Also, the orbit is circular or elliptical, with the result that we have six
possible satellite geometries, namely, (z) equatorial circular, (ii)
equatorial elliptical, (iii) polar circular, (iv) polar elliptical, (v)inclined
circular, and (vi) inclined elliptical.
10.5. Putting a Satellite in the correct orbit
Putting a satellite into the correct orbit involves many operations
related to moving the satellite from the initial orbit into which the
satellite is usually injected by the launch vehicle ta the desired orbit.
This operation is particularly complex in case of geosynchronous satel-
lites where the satellite is first put in a geostationary transfer orbit.
 MICROWAVES AND RADAR
N

Fig. 10.5
The satellite is taken to its final parking slot in steps with the help
of manoeuvres. Sometimes, the operation involves an intermediate
orbit too, between the transfer orbit and the geosynchronous orbit.
Once the satellite is in the correct geosynchronous orbit, that orbit
must be maintained by so-called station keeping procedure.
There are three major processes which affect a satellite in
geosynchronous orbit. These are changes in orbital inclination under
SATELLITES AND SATELLITE COMMUNICATION 499

gravitational attraction of sun and moon, alterations of the semi-major

axis and the mean longitude under the effect of earth's gravitational
field, and finally the changes of eccentricity due to solar radiation
pressure.
Orbital corrections are a crucial feature in a space mission involv-
ing applications satellites. Satellite orbits may have to be corrected for
inexact orbital injection by the launch vehicle for drift in orbital
parameters, or for transferring it from a temporary parking orbit to the
desired one. The corrections are affected by firing a rocket a t the correct
time and with correct impulse.
10.6. Different Satellites - Orbital Requirements
Geosynchronous equatorial orbits are ideally suited for providing
communication and broadcast services. The orbit is usually defined by
the longitude of the satellite's position, for example, so many degrees
west of Greenwich meridian. A set of three satellites can provide global
coverage Fig. (10.6).

m Geostationary orbit

Fig. 10.6
Domestic requirements for communication at high northern
latitudes for which satellites in geosynchronous orbits were originally
considered to be unsuitable have largely influenced the development of
communication satellites in Russia and other countries of erstwhile
Soviet Union. A spacecraft in highly inclined and highly eccentric orbit
offered a more elegant solution.
Molniya orbit, the name given td this orbit, was referred to in an
earlier paragraph. These orbits typically have an apogee a t 40,000 kms
and a perigee at 500 kms and inclined a t 63" to the equatorial plane.
The satellites in these orbits appear to linger for several hours a t a time
over a large part of earth's surface while they are moving slowly towards
and away from the apogee. Conversely, only a few hours will elapse as
they pass rapidly through perigee.
SOP MICROWAVES AND RADAR
Operational requirements of remok sen ing missions are dif-
ferent. On one hand, the viewing geometrq mue t remair. the same, the
orbit should be as circular as possible ~vitheccentricity tjjically less
than 0.002. The altitude selection i c a :ompromise between providing
images over most of the earth's surface anu obtaining good spatial
resolution.
To be able to compare images of the same area taken in different
seasons, and the account for the sensitivity of the cameras to solar light
reflected from earth's surface and atmosphere, t h lighting
~ conditions
have to remain constant. For this reason, remote sensing satellites are
generally put in a sun synchronous orbit. The orbital nodes maintain a
near constant solar time. The satellite always crosses the equatorial
plane at the same local time.
10.7. Satellite Stabilisation
Spin stabilised and three-axis stabilised satellites
The altitude of a satellite needs to be stabilised as for high
communications efficiency, directional antennae pointing at the earth
must be used.
In a spin stabilised satellite, the body of the satellite spins at
30-100 rpm about the axis that is perpendicular to the orbital plane.
The antenna system is generally despun. That is, it is located on a
platform that is spinning in the opposite direction with the result that
it produces a beam that is stationary relative to the earth. A reference
for the control system is usually obtained primarily by IR earth sensors
supplemented by sun sensors. Antenna pointing accuracy off 0.2" or
better is obtained through antenna despin control electronics and by
occasional adjustments to the direction of satellite spin axis.
In 3-axis body stabilisation, there is an internal momentum wheel
with axis perpendicular to the orbital plane. Control about pitch axis is
through wheel's drive motor electronics while control about yaw and
roll axes may be by gimballing the wheel, or by use of hydrazine
monopropellent thrusters to correct axis direction.
Spin stabilised satellites have body mounted solar cells. Three-axis
stabilised satellites using extendible arrays are rotated so as to always
face the sun. A disadvantage of any extendible array for synchronous
orbit missions is the inability to provide power during transfer orbit
when the array is still stored.
10.8. Orbital Parameters
The satellite orbit could be'circular or elliptical. Its characteristic
parameters are governed by Kepler's laws stated below.
( a ) The orbit of a satellite is in general an ellipse with the center
of earth being at one of the foci of the ellipse as shown in Fig.
10.7.
( b ) The line joining the center of earth and the satellite sweeps
equal areas in equal intervals of time.
SATELLITES AND SATELLITE COMMUNICATION 50 1

(c) The squares of orbital periods of two satellites have the same
ratio as that of the cubes of their mean distances from the
center of the earth.
Based on these three laws, we can write expressions for important
orbital parameters of satellites orbiting earth. Derivations of these
expressions are not included as that would be beyond the scope of the
book.

Fig. 10.7
Referring to Fig. 10.7.
Apogee is the point in the satellite orbit farthest from the center
of earth.
Perigee is the point in the satellite orbit closest to the center of
earth.
Orbit eccentricity ( e )is a measure of the displacement of the center
of the orbit from the center of the earth. It is defined as the ratio of the
distance between the center of the ellipse and the center of earth to the
semi-major axis of the ellipse. Orbit eccentricity can also be expressed
in terms of apogee and perigee distances as
Apogee - Perigee
Eccentricity, e =
- + Perigee
Apogee -
Here apogee and perigee represent the distances of apogee and
perigee points from the center of earth.
Other similar expressions are :
Apogee - Perigee
Eccentricity, e =
2a
where a = semi-major axis of the ellipse
Apogee + Perigee
Also, a =
2
Apogee = a (1+ e)
and Perigee = a (1- e)
Orbital Period a n d Orbital Velocity. Orbital period i.e. the
time taken to complete one rotation, (T)is given by :
502 MICROWAVES AND RADAR

where p = GM
G = Gravitation constant = 6.67 x 10-'I ~ - m ~ / k ~ ~
M = Mass of earth = 5.98 x loz4 kg
Orbital velocity (V) can be computed from :

where r = distance of satellite from the center of earth.

It is clear from the above expression that the instantaneous orbital
velocity keeps changing in case of an elliptical orbit due to continuous
change in the distance of the satellite from the center of the earth. This
is a general expression that can be used to compute the satellite velocity
at any given point in its elliptical orbit. For a circular orbit, the
semi-major axis is the same as the radius ( r )which reduces the above
expression to
V=..kL/r
Latitude and Longitude drift i n Inclined Orbits
In case of inclined synchronous orbits, both latitude and longitude
undergo a drift which is a function of inclination angle. This orbit
inclination in effect gives the satellite and apparent movement in the
form of 'figure eight' with the maximum deviation in latitude from
equator given by :

where Lax = maximum latitude deviation in degrees

i = angle of inclination in degrees
Maximum deviation in longitude from the ascending node is given
by (for i < 5 " )
12

where Ymax= maximum longitude deviation from ascending node

i = angle of inclination in degrees
The orbital inclination can be corrected by applying a velocity
impulse perpendicular to the orbital plane when the satellite passes
through the nodes (Nodes are points of intersection of orbital plane and
equatorial plane). For a given inclination angle (i),the required impulse
amplitude is given by :

Azimuth and Elevation Angles

The azimuth angle is defined as the angle produced by intersection
of the local horizontal plane and the plane formed by satellite, earth
station and earth's center with the true north. The azimuth angle,
SATELLITES AND SATELLITE COMMUNICATION 503

depending upon the location of the earth station, can be computed

from :
For Northern hemisphere
If the earth station is towards west of satellite, then

If the earth station is towards east of satellite, then

For Southern hemisphere

If the earth station is towards west of satellite, then
A =A'
If the earth station is towards east of satellite, then
A = 360" -A'
1 tan 1 0 s - 01, I
where A' = tan-
[ sin 01
0s = Satellite longitude
1
0~ = Earth station longitude
01 = Earth station latitude
The elevation angle ( E ) is defined as the angle produced by the
intersection of the local horizontal plane and the plane constituted by

Fig. 10.8
504 MICROWAVES AND RADAR

satellite, center of earth and earth station with the line of sight between

r
the satellite and the earth btation. It can be computed from :
'Rc co"""" 0s-O L ~ -
E = tan-' ~i
Re sin LCOS (XIS 01 cos 1 OS - BL I

I 0 s - OL I )
Earth Coverage Angle and Slant Range
61 cos
1
Rsfer to Fig. 10.8. Earth coverage angle (201) is a function of
elevztion ~ n g l e(E). It can be ccmputed from :

2 u =.2 sin- ' I-Re COSE)

(Re + H
where Re r= radius of earth
H = height of satellite above earth's surface
Maximum earth coverage anple will be for E = C) which gives

2 ~ = 2msin-'~ -
(Re? H ]
The slant range ( d )is given by :

d 2 = ( ~ e + ~ ) 2 + ~ e 2 - ~ ~sin
e(l~
L
I
%+es ii n - ' - [
Re cos E)]
Re+H

Problei.~10.1 :Determine the orbital velocity of a satellite moving

i n a circular orbit a t a height of I50 km above the surface of earth given
that gravitation constant, G = 6.67 x 10-'I ~ - m ' / kg2, mass of earth,
M = 5.98 x loz4kg, radius of earth, Re = 6370 km.
Solution :The orbitai velocity (V ) is given by
V = dp/(R + H)
where p = GM= 6.67 x lo-'' x 5.98 x loz4
= 39.8 x 1013N ~ ~ A ~
R = 63'70 km
H=l5O km
V = 439.8 x 101"(6370 + 150) x lo3
= 7.813 km/s

Problem 10.2 :A satellite moving i n a n elliptical eccentric orbit

has the semi-major axis of the orbit equal to 16000 km. If the difference
between the apogee and the perigee is 30000 k m , determine the orbit
eccentricity.
Solution : Apogee = a (1+ e )
Perigee = a (1-e)
where a = semi-major axis of the ellipse
e = orbit eccentricity
SATELLITES AND SATELLITE COMMUNICATION 505

Apogee - Perigee =a ( 1 + e ) - a ( 1- e )

Apogee - Perigee
or Eccentricity, e =
2a

Problem 10.3 : Satellite - 1 i n a n elliptical orbit has the orbit

semi-major axis equal to 18000 km and Satellite - 2 i n a n elliptical orbit
has semi-major axis equal to 24000 km. Determine the relationship
between their o ~ b i t aperiods.
l
Solution :'The orbital time period (T)is given by
T=2r ja3/11
where p=GM
G = earth's gravitation constant
M = mass of earth.
a = semimajor axis of ellipse
If (al)and (a2)are the values of the semi-major axis of the elliptical
orbits of the satellites 1and 2, (Ti)and (T2)are the corresponding orbital
periods, then
T1=2n=
T2 = 2n
Tz/Ti = ( a ~ / a i ) ~ ' ~

3/2
= (4/3) = 1.54
Thus orbital period of satellite-2 is 1.54 times the orbital period of
satellite-1.

Problem 10.4 : Determine the magnitude of velocity impulse

needed to correct the tnclination of 2 " i n th& satellite orbit 35800 km
above the surface of earth given that radius of earthh = 6364 km, mass
of earth = 5.98 x kg and Gravitation constant = 6.67 x lo-''
~-m'/k2.
Solution :The magnitude of velocity impulse is given by :
tan i
506 MICROWAVES AND RADAR

= 439.8 x 1013/42164 x lo3 tan 2"

= 107 m/s.

10.9. Eclipses
A satellite is said to be under the influence of an eclipse when it
does not receive solar energy which is the source of power for various
subsystems onboard the satellite. The duration and periodicity of the
eclipses therefore have an important bearing on the ability of the
satellite to provide uninterrupted service.

Fig. 10.9
First type of eclipse is the solar eclipse which occurs when the
earth's shadow passes across the satellite as shown in Fig. 10.9. Such
an eclipse occurs on forty four nights each during autumn and spring.
The worst is faced during autumn and spring equinoxes occurring
around September 21 and March 21 respectively. On an equinox, the
inclination angle of the earth's equatorial plane with the direction of
the sun is zero. The duration of the eclipse during an equinox is about
65 minutes. Eclipses on other days have a shorter duration.
The other eclipse called the lunar eclipse occurs when the moon's
shadow passes across the satellite and its effects are similar to those of
a solar eclipse. Its occurence is only once about every 29 years. The
eclipse is diagramatically illustrated in Fig. 10.10.

Fig.10.10
Other than eclipses, another serious form of solar interference
occurs when the satellite passes directly in front of the sun. Due to very
high temperature of the sun, it is a powerful source of noise. This
SATELLITES AND SATELLITE COMMUNICATION 507

effectively blots out transmission from the satellite for a period lasting
about ten minutes for five consecutive days twice in an year.
One of the solutions to provide uninterrupted service during
eclipse periods, which is also the standard solution, is to have storage
batteries which are continuously charged by solar cells during the
period solar radiationis availab1e:~hesebatteriesthen provide standby
power during eclipse period. But the battery capacity required to keep
all the transponders operational may impose a weight penality which
may not be economically justifiable. This often forces the satellite to
function a t a reduced transponder capacity and maintaining only the
priority service. Also, if the satellite is positioned slightly towards the
west of the area that i t services, the eclipse would occur at a time when
the local time on the earth beneath the satellite is close to midnight.
This is the time when data and telephony traffic is low and one can take
advantage of this fact when the satellite is forced to operate at a reduced
transponder capacity during eclipse periods.
The problem of solar interference occurring during the phase when
the satellite passes directly in front of the sun can be resolved only by
switching the transmission channels to another standby satellite before
this interruption occurs. In practice, most communication satellites
have a duplicate satellite in orbit in order to ensure continuity of service
during such interruptions or other serious equipment malfunction.
The angle of inclination of the equatorial plane with the direction
of sun, varies as a function of time and can be computed from :
2nt
i,(t) = 23.4 sin -
T
where T = 365 days
The maximum shadow angle occuring a t equinoxes can be com-
puted from :

\ /
where Re = earth's radius
H = height of satellite above earth's surface
Problem 10.5: Determine the maximum shadow angle that occurs
at equinoxes for a satellite orbiting in a circular equatorial orbit at a
height of 13622 k m above the surface of earth. Assume earth-sradius to
be 6378 km. Also determine the maximum daily eclipse dura.tion.
Solution : Refer to Fig. 10.11
The radius of circular orbit = 13622 + 6378
= 20,000 km

Maximum shadow angle, = 180' - 2 cos-I (%O

-- )
 MICROWAVES AND RADAR

Fig. 10.11

(Pmax = 180' - 2 COS-' 0.3189

= 180"- 2 x 71.4
= 180"- 142.8
= 37.2"
Maximum daily eclipse duration
= x 24 hours

= 2.48 hours

Problem 10.6 : We know that the first day of eclipse before an

equinox and last day of the eclipse after a n equinox correspond to the
relative position of the sun so that the sun rays tangent to the earth pass
through the satellite orbit. Determine the total time from the first day of
eclipse to the last day of the eclipse for the satellite orbit of Problem - 5.
Also determine the same for a geostatio~aryorbit at a height of 35786
k m above earth's surface.
Solution :The angle of inclination ie ( t ) between the equatorial
plane and the direction of sun as a function of time during the time
period of earth's one rotation around sun is given by
2rc t
i e (t)= 23.4 sin -
T
where T = 365 days
23.4" is the maximum inclinxtion that occurs on summer and
winter solstices.
For a circular orbit of 20,000 km radius, I) = 37.4".Therefore, the
time from first day of eclipse to the equinox is given by substituting
ie ( t )= 18.7"(= 37.4"/2)
SATELLITES AND SATELLITE COMMUNICATION 509
CLnt
18.7 = 23.4 sin -
365
365
t =--
2n
sin
-1 1 8.7
(23.4 1
365 x 53 x -
-- n:
- = 53.8 days
2n 180
For the geostationary orbit,
= 180" - 2 cos-'
6378 +. 35786

= 22.13 days

10.10. Components of a Satellite Communication

System
The essential components of a satellite communication system
include :
( a ) a satellite capable of receiving signals beamed at i t from the
earth station, amplifying them and doing frequency transla-
tion before retransmitting the same back towards earth for the
intended users.
(b) transmission path comprising both up-link from the earth
station to the designated satellite as well as the down-link
from satellite to the intended users.
( c ) earth station equipped suitably to transmit and receive signals
to and from the designated satellite.
Components (a)and (b) together are grouped under the heading of
Space Segment while component ( c )is called the Earth Segment. Fig.
10.12 shows different components of a satellite communications set-up.
The space segment and the earth segment are described in a little more
detail in the following paragraphs.
510 MICROWAVES AND W A R

~ransrnittin/
Earth

Fig. 10.12
Space Segment
Space segment as said earlier comprises of the satellite and the
transmission path. The satellite does the seemingly simple task of
relaying the signal received by it after carrying out necessary amplifica-
tion and frequency translation. However, in order to do its job satisfac-
torily, it has to meet very stringent performance standards in terms of
both technical as well as environmental specifications. The important
functional subsystems of a communications satellite (or for that matter
any satellite because it is primarily the payload that differentiates a
communications satellite from satellites intended for other applica-
tions) include the following :
(a) Source of energy
( b ) Power generation and distribution subsystem
( c ) Stabilisation subsystem
(d) Antennae
( e ) Thrust subsystem
(f) Temperature control subsystem
(g) Transponders
( h ) Telemetry, Tracking and Command subsystem

The antennae, the transponders and associated electronics

together constitute the communications payload.
Source of Energy : The satellite derives the electrical power
required for operation of various subsystems from solar energy. An
array of solar cells distributed either around the body of the satellite
(in case of spin stabilised satellites) or on the solar panels (in case of
3-axis stabilised satellites) provide the required opto-electronicconver-
sion.
SATELLITES AND SATELLITE COMMUNICATION 511

Power Generation a n d Distribution Sub-system : The sub-

system generates electric power in terms of required voltages and load
currents for various functional subsystems from available solar energy.
It also contains charging units for the storage batteries. The storage
batteries are continuously charged from the solar energy driven
chargers and are then used to provide electric power to satellite com-
ponents during the periods when solar energy is not available.
Stabilisation Subsystem :The stabilisation subsystem ensures
that the satellite remains in a fixed orientation with reference to its
orbit and that its antennae always point in the right direction. Satellites
are either spin stabilised or 3-axis stabilised. The two techniques were
briefly described in an earlier paragraph.
Antennae :The satellite may use the same or different antennae
for reception and re-transmission. One of the important requirements
of a satellite antenna is to produce the desired shape of the illuminated
area on the earth beneath the satellite what is known as satellite
footprint. Most communications satellites nowadays use several anten-
nae of various sizes, shapes and configurations, so as to produce one or
more of the radiated beams namely omnidirectional, global (or earth
coverage), hemilzone and spot beams. All satellites use an omnidirec-
tional antennae followinginjection of satellite into parlung orbit before
it is positioned in the final designated slot. A non-directional antenna
such as omnidirectional antennae provides communication between the
satellite and the control station until satellite's directional antennae
are fully deployed and oriented in the correct direction.
The earth coverage antennae produce beams with an angle of
17.34",the same as the angle subtended by the earth. The hemilzone
antennae produce beams to cover a specific zone. They can be so
designed as to produce the desired foot print. The spot beam antenna
generate much narrower beams, They are used to concentrate energy
on specific areas of high traffic.
Thrust Subsystem :The thrust subsystem is used to carry out
adjustments to the satellite's orbit, orbital position and altitude. It
comprises of a set of on-board propulsion units along with their fuel
tanks and associated firing and control circuits.
Temperature Control Subsystem :The purpose of temperature
control subsystem is to optirnise heat dissipation by balancing the
temperature distribution across the exterior and within the interior of
the satellite.
Transponders :The transponder subsystem receives the signal
beamed a t it from the transmitting earth station on the up-link, does
the required input filtering, downconverts to a different frequency
(down-link frequency) and amplifies it to the desired level before it is
retransmitted to the intended receiving station. Based on the up-
linWdown link frequency being handled, transponders are categorised
as C-band trmsponders, extended C-band transponders, ku-band
transponders, ka-band transponders and so on.
512 MICROWAVES AND RADAR

Telemetry, Trackingand Command Subsystem :The purpose

of telemetry, tracking and command (TTC)subsystemis to continuously
transmit to ground stations information on vital parameters such as
those I 4ated to satellite's orbital position and attitude and also those
related to the performance specific; tions of various subsystems. In
addition, this subsystem also serves the purpose of receiving commands
from ground stations for ca~ryingout -:ertain corrective manoeuvres.
The TTC system is very important both during orbital injection and
positioning phase and subsequently throughout the operational life of
the satellite. The subsystem has its own dedicated radio link and its
own omnidirectional antenna. The omnidirectional antenna is impor-
tant during the launch am! pusitioning phase. The satellite may con-
tinue to use the same link even during the operational life though there
is an increasing tendency to switch the TTC operations to one of the
transponders and its associated antennae after the initial orbital injec-
tion and positioning phase is over.
Transmission Path :'l'ransmission path can also be considered
as a part of the space segment in additim to the satellite. Transmission
path due to the attenuationit imparts LO the signal and the interference
that it causes to the signal has a great bearing on the performance of
overall satellite communication systrtm. The strength of a radiated
signal is inversely proportional to the square of the distance. As a result,
in a geostationary satellite that is abont 36000 kmfrom earth's surface,
the received signal at the earth statiorl may be cnly a few pico watts.
The attenuation of the signal is becmae of two main reasons. The first
source of loss is the space loss whic.'~is caused by the antennae not being
100 percent efficient. The second source, which is of course the major
source of loss, is the attenuation of s i s a l in the atmosphere. The main
factors responsible for atmospheric losses include presence in the at-
mosphere of oxygen in a free molecular form, uncondensed water
vapour, rain, fog, snow, hail anc! free electrons.
Earth Segments
The earth station and associated infrastructure constitute the
earth segment. The earth station transmits to and receives from a
satellite. An earth station may be located on a ship at sea or even on an
aircraft. Major subsystems of an earth station include :
(a) Transmitter
( b ) Receiver
(c) Antenna
(d) Tracking equipment
( e ) Terrestrial interface
There are three main categories of earth stations namely
(a) Transmit and Receive earth stations
( b ) Transmit only earth stations
(c) Receive only earth stations
Transmit and Receive earth stations are encountered in two way
communication systems. Receive only stations are used mostly direct
SATELLITES AND SATELLITE COMMUNICATION 513

TV broadcast satellite systems and CATV systems. Transmit only

stations are used in data collection systems.
The transmitter converts the signal to be transmitted to the
up-link frequency with proper encoding and modulation. The signal is
then amplified and directed to the appropriate polarisation port of the
antenna feed.
Different components of the earth station receiver include Low
Noise Amplifier (LNA), Down converter, Demodulator, Decoder and
Base band signal processing circuitry. One thing that is common to and
is of utmost importance to all earth stations is the need to achieve a low
system noise temperature in the receiving channel.
Antenna is the most visible part of the earth station. Earth station
performance to a great extent depends uponits antenna characteristics.
The antenna should have a high directive gain, low noise temperature
and a radiation pattern that does not exhibit a large side lobe level. The
commonly used antenna types are the paraboloid antenna with focal
point feed and the cassegrain antenna. Phased array antennas and
Horn antennas are also used.
Tracking and pointing subsystem is used to track the satellite and
point the antenna beam accurately to the satellite in both transmit and
receive modes. The basic earth station comprising the transmitter, the
receiver and the antenna merely provides an access to the high frequen-
cy satellite link. For the link to be usable by the ultimate end users and
for the satellite capacity to be utilised in a cost effective and efficient
manner, the earth station must be equipped with necessary infrastruc-
ture to provide an interface with terrestrial communication services.
10.11. Multiple Access Systems
The multiplc satellite access systems allow geographically dis-
persed locations to access and share satellite's capacity. In fact one of
the major advantages of communications satellites over terrestrial
systems is their ability to link various earth stations thereby providing
single point to multipoint communications. This implies that a satellite
transponder should be possible to be accessed by many earth stations.
Therefore,it is necessary to have techniques for having multiple access
to the transponder capacity. Commonly used multiple access schemes
include :
(a)Frequency Division Multiple Access (FDMA)
( b )Time Division Multiple Access (TDMA)
(c) Code Division Multiple Access (CDMA)
(dl Demand Assignment Multiple Access (DAMA)
Frequency Division Multiple Access (FDMA)
It is the earliest and still the most common form of all the multiple
access techniques of satellite communications. Each of the earth sta-
tions with in the satellite's footprint transmits one or more carriers at
514 MICROWAVES AND RADAR
different center frequencies. Each carrier is assigned a frequency band
with a small guard band to avoid overlapping of adjacent carriers. The
satellite transponder receives all the carriers within its bandwidth, does
the necessary translation and amplification and retransmits them back
to earth. Different earth stations are capable of selecting carriers
containing messages of their interest. Two FDMA techniques are in
operation today. One of them is multichannel per carrier where the
earth station frequency division multiplexes several channels into one
carrier baseband assembly which then frequency modulates an RF
carrier and transmitted to FDMA satellite transponder. This technique
is called FDM-FM-FDMA. In the other technique called single channel
per carrier (SCPC), each channel modulates a separate RF carrier and
then transmitted to FDMA transponder. Advantages of FDMA include
simplicity of each station equipment and the fact that no complex timing
and synchronising techniques are required. Disadvantages include
likelihood of intermodulation problems with its adverse effect on signal
to noise ratio.
Time Division Multiple Access (TDMA)
Time Division Multiple Access (TDMA) is a technique in which
different earth stations in the satellite's footprint making use ofsatellite
transponder use a single carrier on a time division basis. Different earth
stations transmit traffic bursts in a periodic time frame called TDMA
frame. Over the length of a burst, each earth station has the entire
transponder bandwidth at its disposal. The traffic bursts from different
earth stations are synchronised so that all bursts arriving at the
transponder are closely spaced but do not overlap. The transponder
works on a burst at a time and retransmits back to the earth sequence
of bursts. All earth stations can receive the entire sequence and extract
the signal of their interest. Advantages of TDMA include absence of
intermodulation problems as the transponder has to deal with one
carrier at a time. Disadvantages include requirement of complex and
expensive earth station equipment and stringent timing and
synchronisation requirements. TDMA is suitable for digital transmis-
sions only.
Code Division Multiple Access (CDMA)
In Code Division Multiple Access (CDMA), the entire bandwidth
of the transponder is used all the time and signals from different
stations are encoded in such a way that information from an individual
transmitter can be recovered only by a properly synchronised receiving
station equipped with the matching decoder. Thus while the capacity
of communication is increased by multiple acces_s,the secrecy of trans-
mission is maintained. CDMA is quite similar to spread spectrum
communication techniques and has potential applications in military
communications using satellites.
Demand Assignment Multiple Access (DAMA)
While the conventional Frequency Division Multiple Access
(FDMA),Time Division Multiple Access (TDMA) and Single Channel
SATELLITES AND SATELLITE COMMUNICATION 515

Per Carrier (SCPC) multiple access techniques, in their basic form,

allocate fixed amounts of capacity to users, the use of DAMA technique
enables the allocated capacities to be varied dynamically according to
individual demands. Demand assignment is used in conjunction with
SCPC and TDMA systems.
10.12. Satellites for Communication Services
In the field of communication applications using satellites, the
main application areas include satellite television, telephone services
via satellite, and data communication services. The underlying prin-
ciple, the application potential, and the contemporary status of the
different communication related satellite applications are discussed in
the following paragraphs.
Satellite television
Satellite television is one of the most widely used and talked about
application areas of satellites. In fact, if 'satellite' has become a
household name today, i t is largely due to its television broadcast
capability.
Satellite television refers to the use of satellites for relaying TV
programmes from a source point from where these originate to a large
geographical area. Different configurations are available for use to
implement satellite TV but, irrespective of the configuration used, the
satellite basically receives a certain telecast beamed upwards known
as uplink. The signals are then reflected back towards the earth without
any change, except for carrier frequency translation and amplification,
and is called the downlink. This amplitude and frequency translation
job is done in the satellite's payload known as transponder. A
transponder translates the received signal to a new frequency band and
then amplifies them for transmission back to earth. The S, C, X, and K
bands, as shown in Table 10.3 are exploited for satellite based ,com-
munication services.
Single point t o multipoint connectivity
The single point to multipoint connectivity feature of satellite
communication, as depicted in Fig. 10.13, is the basis of one of the most
common satellite TV configurations. There is one transmitting station
and many receive only stations designated as TVRO stations. This
configuration makes use of the inside coverage area of the satellite's
footprint. The satellite that makes use of this type of configuration is
known as direct broadcast satellite. The programme to be .telecast
beams up at the satellite via uplink from the studio (Fig. 10.14). In the
arrangement of Fig. 10.14, it is, however, assumed that the studio has
its own earth station. In case the earth station with the uplink facility
is remotely located, the programme is usually sent to the earth station
over a microwave terrestrial link. The down link is received at each TV
station by its own receive only station from where it is transmitted to
the viewers through the local TV channel.
516 MICR.0WAVES AND RADAR

Fig.10.13

Fig. 10.14
 Table 10.3

There are instances where place of action is remote such as in live

telecast of sports events. The signal picked up by cameras in such cases

PROQRAM
DISTRIBUTION
SATELLITE

RECEIVE SITES

Fig. 10.15
is transmitted to the TV station with an uplink facility if the distance
involved is few tens of kilometres. If the distance involved is great, a
mobile earth station in the formof avan is posted at the site. The signals
are then transmitted to the TV station over another satellite link on
point-to-point connectiv$y basis. The programme reaches the viewers
via another satellite link on single point to multipoint connectivitybasis
(Fig. 10.15). A number of countries today have their own domestic
satellites that are used for direct broadcast purposes. Some of the
popular direct broadcast satellite systems include INSAT (India) (Fig.
10.16) AUSSAT (Australia). TV-SAT (Germany), TDF-I (France) and
STC (USA).
Cable television
The concept of cable television originated from the need to have a
common receiving antenna for a large number of houses. The signal
received by this common antcnna could be distributed to several houses
through a network of coaxial cables. Such a system is ideally suited to
large buildings having hundreds of apartments, eliminating the need
for each house to have its own receiving antenna (Fig. 10.17).
518 MICROWAVES AND RADAR
fi

Fig. 10.16
-DISH ANTENNA

Fig. 10.17
The concept of cable TV which in the past meant community
antenna has assumed a much bigger role in the present day cable TV,
the cable TV operator's premises is a sort of receive-only earth station
capable of receiving telecast of one or more than one satellites either
with the help of a single dish-antenna with multiple feeds (with each
feed so aligned as to receive a telecast from a different satellite) or by
having more than one dish antennae. The cable operator routes the
received signal or signals simultaneously over a network of cables on a
single point to a multipoint connectivity basis to a large number of
subscribers. The operator also uses this network to transmit other
recorded programmes that he has received from other sources.
In fact, we are entering an era where each house-top will have its
own antenna dish capable of receiving TV programmes directly from
satellites. TV receivers with built-in decoders have arrived in the
market place. These TV sets' with an appropriate dish antenna, would
be capable of direct satellite TV reception.
SATELLITES AND SATELLITE COMMUNICATION 519

Telephone services
Today more than 80 per cent ofinternational telephone traffic goes
via satellite, which like any other telecommunication service,serves the
purpose of a repeater station. When compared to terrestrial link, the
satellite link is particularly useful where the distance involved is large
or when the region to be covered is sparsely populated or has a difficult
geographical terrain.

SATELLITE

it
TELEPHONES
8
TELEPHONE

Fig. 10.18
With the advances in digital technology, the echo problem as-
sociated with telephone services via satellite has been overcome using
echo cancellation techniques. Fig. 10.18 shows the typical set-up of an
international telephone service using a satellite link.
Satellites are playing a vital role in the global communication set
up. While on one hand INTELSAT-series of satellites use geostationary
satellites to provide communication services on the global level, arecent
system conceived by Motorola and known by the name of Project
Iridium, will make use of satellites placed in lower earth orbits. A total
of 77 satellites are arranged in a distributed architecture with each
satellite carrying ( W 7 ) of the total system capacity (Fig. 10.19). The
system is intended to provide a variety of telecom services on the global
level.
Data communication services
Satellites have also revolutionised the data communication ser-
vices. The purpose of a data communication serviceis to link computers
and other data processing facilities located thousands of kilometres
apart. Terrestrial communication channels, used for data communica-
tion, have a relatively slower data transfer rate, higher cost, lower
reliability and a lower throughput capability. Satellite links are attrac-
tive for data communication, particularly because of much higher data
transfer rate, higher reliability and much higher performance to cost
ratio.
520 MICROWAVES AND RADAR

One of the most common data communication applications imple-

mented by a satellite is data broadcasting. Data broadcasting makes
use of a single point to multipoint connectivity configuration.
In a typical data broadcasting set-up (Fig. 10.201,data from a
central facility is routed to the earth station with the uplinking
capabiliQ via another set-up called data packet assembler which
receives the data to be broadcast, organises it into packets and places
appropriate address bits at the beginning of each packet. Each data
pack9tiis thus self-contained with source and destination addresses,
and therefore can be routed in any way over the sateUite link and finally
to r terrestrial packet switched network.

Fig. 10.19.
Each data packet i s even coded so that only authorised users can
have access to the information. Full access to all the data is also
provided in case of certaifi services such as news service. In such cases,
data beamed up at the satellite is retransmitted towards the earth to
be received by RO terminals. The RO terminals have digital processing
capabilities to identify data pocket addressed to them. The respective
RO terminals receive data meant for user terminals linked to them for
subsequent delivery. The user terminals (shown as a PC in Fig. 10.20)
are usually linked to the central facility over telephone lines to enable
them make a request for broadcast of any data of interest.
Another type of data communication service is the one provided by
an interactive data communication network. In an interactive set-up
(Fig. 10.211, the user terminals are not only able to respond to the
SATELLITES AND SATELLITE COMMUNICATION 52 1

I
RECEIVE-ONL
DATA T E R M I N I
DATABASE

DIALdJP LINE F O R
DATA R E O U E S T S

Fig. 10.20
central facility, they can also exchange information between one
another. Such a network is best implemented using VSATs (very small
aperture terminals) with each VSAT terminal supporting a large num-
ber of user terminals.

mk
PROCESSOR
TYPICAL REMOTE
LOCATION

Fig. 10.21
In the set-up shown in Fig. 10.21, the remote station can transmit
information to the central facility using the satellite that does the data
broadcast. There is no need for a terrestrial link for the purpose. The
data concentrator multiplexes the data coming from different user
terminals and sends out a single bit stream to the VSAT. It also
522 MICROWAVES AND RADAR

demultiplexes the data received from the satellite for onwards delivery
to prospective user terminals.
10.13. Satellite for Weather Patrol and
Atmospheric Monitoring
The idea of launching satellites in fifties, followed by the success
achieved with the early experimental satellite missions, mainly tar-
geted to carry out some scientificstudies forced the owner countries like
the United States and the former Soviet Union to think of more practical
applications of this marvel of technology that had just then seen light
of the day. The idea of weather patrol in space or that of observing the
planet earth were indeed exciting. The importance of examining the
earth and the atmosphere surrounding it from space with instruments
such as high resolution cameras, radiometers etc. was quickly ap-
preciated by advanced countries. Since then, satellite meteorology and
remote sensing have come alongway and today, the number of satellites
in space for meteorological and remote sensing applications are next
only to communications satellites. These satellites are being used to
provide remote sensing and meteorological data on the global level on
a routine basis.
Meteorological satellites -how they function
The idea of computing the future state of atmosphere from the
mathematical equations which describe its behaviour was first con-
ceived by Dr. L.F. Richardson in 1916. Such calculations, however, could
not be carried out until the advent of modern computers. The accuracy
of a forecast depends not only on the computing power but also on the
quantity and quality of observational data. Weather satellites are
invaluable aids to the analysis of weather. The most familiar and
probably the most widely used satellite products are the pictures,
sometimes showing striking cloud formations. These pictures contain
latent information about cloud type, and also about position, extent and
intensity of frontal depressions, thunderstorms, hurricanes, sea breeze
circulation, and so on. Details of snow cover, sea ice, sea surface
temperatures, cloud top temperatures and winds at different heights
are also extracted regularly from the satellite pictures.
The most important payload of any modern weather satellite is the
radiometer. It is the instrument that is used to obtain the images by
measuring the amount of radiation. The radiometer scans the area to
be picturised, and then builds up the picture in terms of the radiation
amounts it receives from different parts of the area. The radiometer is
usually provided with filters to allow the radiation of known wavelength
range. Measuring in the visible part of the spectrum provides pictures
about the prominent features such as clouds, deserts, snow etc. On the
other hand, measurements in the infrared part of the spectrum can be
used to determine relative temperature of the different features. In-
frared measurements also allow us to take pictures at all times during
the day as well as during the night.
SATELLITES AND SATELLITE COMMUNICATION 523

Fig. 10.22
Pictures are usually obtained in black and white, with black areas
indicating relatively higher temperatures and white areas indicating
relatively low temperatures. High clouds which are cold, therefore,
appear white and lower clouds appear in different shades of grey. Fig.
10.22 is a sample picture taken from a weather satellite which shows
the central part of the world. The black, the white and the parts with
different shades of grey can be clearly noticed from these photographs.
Meteorological satellites-the early developments
Use of satellites for weather patrol and atmospheric monitoring
began with a very small yet significant space borne meteorological
experiment carried out on a small satellite called Explorer-7 launched
by the United States in 1959. It carried onboard a very crude form of
radiometer comprising of two hemispheres, one of them painted black
to absorb all wavelengths and the other painted white to reflect all solar
energy except terrestrial infrared radiation. Since then, satellite
meteorology has come a long way through various phases of advance-
ments in the space techniques for meteorology as well as in the
meteorological payloads.
The TIROS-series (an acronym for Television and Infrared Obser-
vation Satellite), satellites constitute one of the earlier weather satellite
systems followed by ESSA (Environmental Sciences Services Ad-
624 MICROWAVES AND RADAR

ministration) programme, and subsequently followed up by improved

TIROS operational satellites system abbreviated as ITOS. Today, there
are a large number of operational weather satellite systems, with
participation from both developed as well as developing countries. The
METEOSAT from European space agency, Russia's METEOR weather
satellite system, SMS/GOESseries from USA, GMS system from Japan
and India's INSAT system which, of course, is a multipurpose mission,
are some of the international systems other than TIROS, ITOS and
ESSA series.
Meteorological satellites have two major objectives which comple-
ment each other. These are to observe atmospheric phenomena and to
understand the mechanisms responsible to produce them. A major
portion of space research today concerns observation and under-
standing of climatic changes. Recent experiments have laid emphasis
on the study of oceans as they can force the climate. The long term
objective of meteorological research, of course, is the prediction of
climate and its changes whether man-made or natural. This work is of
great economic and socio-economic importance, and is absolutely vital
in the world of tomorrow.
-
Meteorological satelfites different families
There are two broad categories of meteorological satellites. The
satellites belonging to the first category are the ones having a geosta-
tionary orbit. The polar satellites constitute the other class. The
meteorological satellites from the two families complement each other
very well. While the geostationary satellites view a wide band of
latitudes towards the north and south of equator, the satellites in the
polar orbit and having an altitude of 600 to 1600 kms can be used to
repeatedly observe the polar regions. While the geostationary
meteorological satellites are not effective over the polar regions as their
field of view is distorted at high latitudes, low altitudes of polar
satellites, on the other hand, favours a good spatial resolution on ground
and in the atmosphere. TIROS-N series from USA and METEOR series
satellites from Russia are the major meteorological satellite systems
belonging to the class of polar satellites. The major geostationary
weather satellite systems include the Europe's METEOSAT, USA's
SMS/GDES series and Japan's GMS. A netwcrk of satellites comprising
of satellites from the two families can be used to advantage to provide
round-the-clock global meteorological information. One such network
of satellites that took part in the famous Global Atmospheric Research
Programme (GARP) during December, 1978- November, 1979 to im-
prove our knowledge of global processes that bring about changes in
weather and climate is shown in Fig. 10.23. Some 150 odd member
countries of World Meteorological Organisation and International
Council of ScientificUnions contributed to this grand co-operative space
effort. The objective of this joint experiment was to determine the
practical limits for weather forecasting so as to get sufficient data that
would help in designing an appropriate world observation system.
SATELLITES AND SATELLITE COMMUNICATION 525

During t h i s experiment, t h o u s a n d s o f o b s e r v a t i o n s and m e a s u r e m e n t s

SMSGOES

U
METEOSAT

Fig. 10.23

GOES (EAST)
USA

GEOSTATDNARY
ORBIT ?-
I

GMS u GOES (WEST)

(JAPAN) USA

Fig. 10.24
526 MICROWAVES AND RADAR

were made from ships, aircrafts, sounding rockets and balloons to trace
air and moisture movements over every region of earth's surface.
Another network of satellites belonging to the two families and imple-
mented in 1987 is shown in Fig. 10.24.
10.14. Remote Sensing Applications
Observing the planet earth using satellites, designed as remote
sensing satellites, has come of age. What subsequently led to the
realisation of scores of remote sensing applications began with the use
of TIROS weather satellite in early sixties, when the satellite sent
map-like outline of the world beneath the clouds. Today, the remote
sensing satellites find extensive use in monitoring agriculture, forestry,
detection of hidden mineral resources, oil exploration and cartography
to mention some of the prominent ones. While the remote sensing
satellites are being exploited for newer areas of application, there has
also been great emphasis on the development of satellites for earth
observation. Quite interestingly and expectedly, besides the countries
of the developed world, the developingworld countries have also shown
keen interest in this field of satellite usage.
Satellite remote sensing vs conventional methods
Information gathering by conventional methods such as terrain
surveys using aerial photography and statistical studies is expensive
and is also not being updated rapidly enough. This makes it impossible
to achieve a complete and homogeneous survey of an entire country. In
addition, conventional methods are sometimes difficult to employ, par-
ticularly for countries which have inaccessible terrains. Satellite remote
sensing enables homogeneous observations to be made over vast
regions, perhaps thousands of kilometers in one image.
The main advantage of satellite remote sensing is that the whole
earth is accessible to satellite observations, regardless of geography or
climate of the region. Also, these measurements can be repeated peri-
odically, thus allowing changes, if any, to be identified. The repetitive
nature of satellite orbits increases the possibility of obtaining images
unobscured by clouds. Since the orbits of remote sensing satellites are
usually sun synchronous, the repetitive images are obtained at the
same local time so that over a period of four weeks, the angle of the sun
and the shadow remain the same. This makes i t easier t o identify
changes.
In remote sensing using satellites, the repetitive orbital perspec-
tive increases the chances of cloud-free imagery and identifying slow or
radical changes. Satellites can monitor vast areas quickly and much
more economically than the conventional means.
Monitoring agriculture
Remote sensing data is used for several applications such as crop
identification, agricultural crop acreage and field estimation, drought
monitoring and assessment, flood mapping and land cover mapping.
Present-day remote sensing satellites equipped with the state-of-the-
art high resolution cameras are capable of monitoring characteristics
 SATELLITES AND SATELLITE COMMUNICATION 527

like crop yield, optimum plantation and harvest Ymes, land fertility etc.
over much smaller areas.
With the analysis of the data received from the modern remote
sensing satellites, we can have a global food watch, that could help
mankind avoid disastrous food shortages. The difficulties accompany-
ing over or under production can be avoided by accurate forecast of
optimum time for plantation and harvest, mop acreage etc. This is done
by monitoring soil condition and moisture content by keeping an inven-
tory of crops during the growing season and by giving advance warnings
of drought. This type of agricultural watch could also allow an inventory
to be made of tropical areas which could be potentially productive. It
could also give data on fertile and arid regions which could be made
more productive by irrigation.
Management of forests
Forest resources survey and management is another very useful
application of remote sensing satellites. Satellite data has shown clear
advantages in monitoring deforestation and, if necessary, to recom-
mend changes in the cutting pattern. Proper management of forest
resources is particularly attractive for developing countries due to the
need to preserve ecological balance, prevent erosion, silting of dams and
pollution of coastal waters.
In India too, the remote sensing data from IRS-series satellites is
being used for forest resources survey and management. LANDSAT
satellite imagery in fact has been used to monitor programmes for
controlled development of forests and also for monitoring large scale
forest fires and consequent damage.
Cartography
Cartography, which means map-making, is one of the early ap-
plications of remote sensing satellites. Satellites with the help of high
resolution cameras onboard these satellites have not only been used to
correct and update certain features of existing maps in many countries
including USA and Russia, but satellite imagery has also been used to
pp6duce relevant maps for huge construction projects such as building
&ighways, railway tracks or irrigation channels. It has also been used
to map underwater features such as coral reefs that pose potential
+ threats to ships.
In India too, satellite imagery has been used to provide coral reef
mapping for the Gulf of Kuchch, Lakshadeep, Gulf of Mannar, and the
Andaman and Nicobar Islands. Fishery prospects charts have also been
generated using data from IRS-series satellites.
Satellite imagery is so effectivein agricultural monitoring that the
farmland of an entire repion s ~ r e a d
over hundreds of kilometres could
be sorted crop by crop inva maker of a few hours with the processing of
data from satellite being done by high speed computers on ground, the
end result being a computer printed terrain map showing precise
location and area of each crop.
528 MICROWAVES AND RADAR

Oceanography
Remote sensing satellites are being increasingly used to send data
enablixg determination of sea conditions such as sea surface tempera-
ture (SST),wind direction and speed, ocean waves conditions, ice fields
and icebergs, tidal conditions and so on. With more and more sophisti-
cated sensors getting placed onboard ths present day earth observation
satellites, they are likely to get ciata on subjects such as coastal
sedimentation and pollution, conservation and use of fish stocks, ship
routing making use of ocean currents, wave forces for use in the design
of off-shore structures, mapping of polar ice caps, ocean temperatures
and winds for improv~dclimate and weather forecasting.
The satellite data can also be used to predict the behaviour of
icebergs which is very crucial for navigation of ships in cold waters. The
satellite data provides information on sea temperatures, precipitation,
wind and currents, ice thickness and so on.
Exploring oil and mineral resources
Remote sensing satellites are eq~~ally popular with geologists who
often use satellite data to see earth's interlocking features including
large folds and ruptures which give clues to mineral deposits. Transcur-
rent faults, as indicated by satellite d a t ~can
, be used to make searches
for unknown oil reserves. Minerals can be identified by rock colour and
its topographic form.
In India, IRS-satellite data has been effectively used to identify
favourable locations for gold rnincralisation such as Schist belt in the
Gadwal district in ~ n d h i Pradesh.
a The LANDSAT imagery has been
used for selecting site for nuclear power stations and routes for
pipelines.
Pollution monitoring
Remote sensing satellites equipped with instruments like correla-
tion interferometers are being used for mapping those parts of atmos-
phere which have some concentrationof carbon monoxide. By repetitive
scanning over long periods, it is possible to define removal mechanism
which enables prediction of futuristic behavioural pattern of carbon
monoxide for monitoring oil slicks in the sea. It is very difficult to see
such leakage from aircrafts which are restricted in vision to narrow
bands of ocean because of their low altitudes. The satellite imagery has
no such limitation.
The satellite data can be used to not only detect the oil spillage, it
can also be used to access the quantity of spillage, identify its type and
quality which could lead to identification of source of leakage. With
sophisticatedphotography instrumentation such as multispectzalscan-
ners, return beam vidicons (RBV), thematic mappers and other in-
strumentation payloads such as correlation interferometers, the remote
sensing satellites are doing a marvellous job in the areas of oil and
mineral exploration, pollution monitoring, oceanography and terrain
mapping.
SATELLITES A N ~ L L I T COMMUNICATION
E 529

10.15. Scientific and Technological Applications

Prior to the development and successful launching techniques of
satellites, our access to the universe from the ground-basedobservation
posts had been severely limited by the earth's atmosphere. Due to
absorption of a large part of electromagnetic spectrum in the lower
atmosphere, the study of distant planets, stars and galaxies, that is
based on the electromagnetic radiation emitted by these celestial
bodies, had been restricted to a very narrow band of electromagnetic
spectrum.
With the ground-based observation posts, most of the lower fre-
quency radio waves and also the extreme frequency ultraviolet radia-
tion, X-rays and gamma rays are inaccessible. In addition, the studies
have also been severely hampered by bad weather conditions, man-
made pollution and the background heat radiation of the earth itself.
These frequendes, which had been inaccessible till the deployment of
satellites for carryingout these studies, are the ones that have revealed
the truly violent nature of out vast universe.
The clarity, finesse and the depth with which we know the universe
today is largely due to the use of satellites. Satellites equipped with
scientific payloads offer an effective solution to carry out atmospheric
and space studies.
Probing the atmosphere
One of the earlier missions specially designed to study the aerosol
concentrationof the stratosphere was the stratospheric aerosol and gas
experiment (SAGE).The payload (Fig.10.25) used to carry out the study
DEPLOYABLE SOLAR
ARRAYS

ITA ANTENNA
TRY
SAGE

Fig. 10.25
630 MICROWAVES AND RADAR

was launched aboard Satellite Applications Ex- iorer Mission-B. The

SAGE experiment was the result of fear and concern that a large
number of nations had industrial pollution that could lead to an exces-
sive build-up of aerosols and gas in the stratosphere, causing adverse
changes in the climate.
The stratosphericregion that extendsfrom about 12kms to 50 kms
above the surface of earth is also significant fkom another viewpoint
that is, the ozone in this region protects us on this planet earth from
the harmful effects of the ultraviolet part of the sun's radiation. l'his
region also contains aerosols, small particle or liquid droplets con-
centrated below the ozone. The aerosol concentration does affect the
efficacy with which ozone protects us from ultraviolet radiation.
The Explorer-B satellite that carried the SAGE instruments was
launched in 1979. It operated in 600 km circular orbit. The SAGE
photometer looked at the sun through the stratosphere gas and
aerosols, everytime the satellite entered or left the earth's shadow. The
instrument recorded fifteen sun rises and an equal number of sun sets
in 24 hours. From repetitive observations, the SAGE experiment was
used to determine aerosols and ozone concentrations.
Another significant study was that of magnetosphere carried out
during the period 1976-79where it was designated as the international
magnetosphere study (IMS)for the purpose of determining the effects
of solar activity on the weather patterns, the long-termof effects of solar
winds and the interplanetary magnetic field on the world's climate, and
the relationship between the sun and the ionosphere as regards com-
munications, navigation and ozone levels.
Indian Space Research Organisation (ISRO)in India is also pur-
suing a programme of using satellites to carry out scientific experiments
of relevance to the Indian context. the sucoessful development and
launching of Stretched Rohini Series Satellites (SROSS)has achieved
many objectives in the field of space science.
Space studies
Studying space which includes probing different planets, studying
interplanetary forces, solar wind and solar flare phenomenon, discovery
of new phenomena, to mention a few, is another important mission of
scientific satellites. Both unmanned vehicles as well as manned obser-
vatories have been used in the past to carry out space studies.
It is felt that many observations and studies can be best carried
out by humans in space. Many manned orbiting platforms have been
launched by the USA and the earstwhile Soviets for the purpose. These
include Soviet's Salyut and USA's Skylab. These missions provided a
unique opportunity for solar and other astrophysical investigations.
The Orion-1 stellar spectrograph carried by Salyut-1 and the
orbital solar telescope about Salyut-4 gave scientists a chance to record
solar activity over an c-xtended period with virtually no interruption
due to man-machine interface to catch unusual solar fluctuations.
SATELLITES AND SATELLITE COMMUNICATION 53 1

Hundreds of solar photographs and spectrograms were taken by the

crew manning Salyut-1 during their flight. The Skylab spacc station
also contained no less than eight solar telescopes, five of them being for
X-ray and ultraviolet radiations, two for television solar pictures, and
one for photographing solar corona. As many as 180,000 photographs
were taken during a total of six months occupation of Skylab station.
The photographsincluded the first recordingof the birth of a solar flare.
Another example of the effectiveness of manned observatories and
actual human intervention in space was the Solar Maximum Mission
(SMM)satellite launched in February 1980 to study the sun during the
most active period of its 11-yearcycle. The mission observed solar flares
in ultraviolet, X-ray and gamma rays regions of the spectrum and
measured the sun's total radiation output within 0.1 per cent. The
scientific payload for SMM satellite included gamma ray spectrometer,
ultraviolet spectrometer, hard X-ray spectrometer, hard X-ray burst
image spectrometer, X-ray polychromator, coronagraph, polarimeter
and active cavity radiometer irradiance monitor.
Satellites with scientific payloads have provided significant data
on composition of different layers of earth's atmosphere such as mag-
netosphere, stratosphere, ionosphereetc., and the effects of the changes
in their compositions on weather patterns, communication, navigation
and so on.
Infrared Astronomical Satellite (IRAS)launched in 1983 was the
first to carry out an all sky survey to look for astronomical objects
emitting infrared radiation. IRAS was an international venture with
the participation of the USA, the UK and the Netherlands. As the IRAS
payload responded to infrared part of the electromagnetic spectrum, it
could detect radiation from the stars that could not be seen by telescopes
limited to observations in the visible light wavelengths.
Another astronomical satellite of European origin launched in
1983was the EXOSAT. It was designed to measure position, structure
aspects and speckal and temporal characteristics of X-ray sources. It
made more than 2000 observations, mainly on the physics of neutron
stars and black holes.
New spacecrafts for scientific studies
Hubble telescope, Infrared Space Observatory (ISO), Gamma ray
Observatory (GRO)Advanced X-ray Astrophysics Facility (AXAF),Hip-
parcos, Extreme Ultraviolet Explorer (EUV)snd Soho-Cluster missions
are some of the new spacecrafts designed for scientific studies.
United States Hubble space telescope (HST) equipped with a 2.4
m diameter mirror is capable of viewing stars as faint as 28th mag-
nitude compared with 25th magnitude achievable from ground-based
systems. The main objective of (HST)mission, successfullylaunched by
shuttle in May 1990, was to place the above mentioned 2.4 m diameter
astronomical telescope and its associated instrumentation in low earth
orbit of 600 kms a t 28.5 degrees inclination and to maintain it over a
15-year period as an international observatory.
532 MICROWAVES AND RADAR

Ground-based telescopes rarely provide better than 2 are sec

resolution, except under the very best viewing conditions. The 0.02 arc
sec resolution of Hubble telescope is what has made possible for it to
see much fainter stars, about 20 times fainter than what could be seen
from ground-based telescopes. By another yardstick, HST is capable of
penetrating 14 billion light years.
Gamma ray observatory (GRO)has been another prestigious space
mission by the United States. The advanced X-ray astrophysicsfacility
(AXAF) from NASA, USA is similar to Hubble telescope in design. An
X-raytelescope of exceptionalpower and precision aboard AXAF carried
out spectroscopic studies.
After IRAS, another infrared observatory from European Space
Agency is the Infrared Space Observatory (ISO). With the availability
of the pioneering all sky survey from IRAS, the next logical step was to
examine in detail the sources of infrared radiation. This needed a
telescope which could be held pointing to source for long periods, and
which was equipped with instruments of higher spatial and spectro-
scopicresolution and higher sensitivity. The answer was found in ESA's
ISO.
Soho-Clusteris a multi-satellite project. While Soho was to observe
the sun, its corona and the solar wind, the four cluster spacecraftswere
to study the reactions of the magnetosphere to varying solar conditions.

Fig.10.26
SATELLITESAND SATELLITE COMMUNICATION 533

US'S FREEDOM space station (Fig. 10.26) and Russia's MIR-2

space station are the two most contemporary space missions. Both these
stations would be permanent research facilities. While the Freedom
space station is planned to be assembled in space over a period of three
years using about 20 launches of the manned space shuttle, the MIR-2
space station is to be launched by Energia. This would theoretically
enlarge observation of the visible universe 40 times over the capability
of Hubble telescope. It may also enable us to determine not merely the
dimensions of the universe but also its mass and density.
Materials and life sciences studies
Research into behaviour of materials in microgravity environ-
ments is one of the most recent subjects in space research. Work in this
area began in right earnest once space labs were in orbit around the
earth, and the microgravityenvironment could be obtainedfor days and
even weeks. The massive manned space stations are being used now to
carry out studies related to materials and life sciences in space.
Materials processing in space takes advantage of microgravity
conditions that exist in space. It may be mentioned that true zero
gravity does not exist as gravity extends theoretically to infinity.
Typical value for the gravity is g for the microgravity environment
being referred to.
When the materials are processed in space, the loss of weight
condition has been observed to play a significant role in forming new
substances. With out going into details, it would suffice to say that
processing of materials in space promises to end a whole range of
production problems. Scientists and researchers are confident that new
alloys and composites will be s o w n in space. Microgravity processing
of materials holds a very bright future for the growth of ultra-pure
semiconductor crystals for the electronics industry.
Although silicon of sufficient quality is produced in terrestrial labs
for use in making semiconductors, new materials such as gallium
arsenide, mercury cadmium telluride and tin telluride would present
technological challenges when these materials are grown in space.
The success achievedis due to experiments made under conditions
of weightlessness, where effects of different constituents having dif-
ferent densities are minimised. The semiconductor materials grown in
space could be used in making super high speed and very large scale
integrated circuits.
Study of life sciences under microgravity conditions is also being
developed with great interest. One of the areas of study is the long term
effects of microgravity on human body and its functioning, and another
is to study other effects of gravity as a variable.
10.16. Military Applications
Although, satellites have always had a major military role to play,
both during war as well as peace time, the extent to which satellite
technology has superseded the earlier methods of communication,
534 MICROWAVES AND RADAR

reconnaissance,navigation, metrology etc. is largely responsible for the

ever increasing role of military satellites. Some of the common military
applications of satellites include intelligence gathering and reconnais-
sance, command and communications,navigation, meteorologicalfunc-
tions, providing early warning of impending missile attack and
detectioc of nuclear teets.
Theluxuries ofpossessingmilitary satellites are, however, enjoyed
at present mainly by major military powers.
One of the main functions of military satellites is related to
reconnaissance and intelligence gathering. In this context, military
satellitzs are used to verify the extent and composition of military
forccs, verification of agreed limitations of stratkgic arms and monitor-
ing af u?w military developments.
'Dscoverer'was one of the earlier satellite-seriesin t h s class. Data
from Discoverer, which was basically a photo-reconnaissance satellite,
was i,~strumentalin ending US fears of a massive earstwhile Soviet
ICBM build-up. The Discoverer-series satellites had a re-entry capsule
that could carry the exposed photofilm back to earth.
'Search and find' satellites followed Discoverer series in the USA.
Subsequently, different generations of reconnaissance satellites were
evolved with the advances made in instrumentation and satellite tech-
nology, and also with ever increasing complexity af strategic military
requirements.
Russia and China are the other countries pursuing active recon-
naissance satellite programmes. Russians and Americans have also
tested manned orbital reconnaissance. While Russians have done it in
the military adaptation of Salyut space station (Salyut 6 and 7) and the
MIR space station, Americans have done it under manned orbital
laboratory (MOL)project.
Use of radar aboard reconnaissance satellites emerged in 1980s.
Photograph using radar has the advantage that it is not hampered by
bad weather conditions. Other surveiliance and reconnaissance related
military applications of satellites include the electromagnetic recon-
naissance dealing in location and catalogingof military radio and radar
stations and ocean surveillance.
Navigation is another application. Navigation satellites enable
position f i n g with an accuracy of a few metres. US navy's TRANSIT
series and the more recent NAVSTAR are navigation satellites.
Receivers fix their position through a combination of doppler effect and
accurate orbital details transmitted by satellites.
Command and communications is surely the most important ap-
plication ofmilitary satellites. The satellites meant for this purpose can
provide instant communication between the operational forces and
their national command authorities, irrespective of the distance and
time of the day. DSCS (DefenceSatellite Communication System)is one
such series whose launching began in 1966 and continued till mid-80s.
SATELLITES AND SAIXLLITE COMMUNICATION 535

The other important military satellite systems include 'Seatsat' used

for naval applications. 'SDS' used to communicate with US nuclear
forces in polar regions, Fleetsatcom to provide global military coverage
to US Air force, Army and Navy, NATO-I11 series to provide rapid and
secure communications among NATO members and Britain's Skynet-
IV to provide links among ships, mobile terminals on land and control
centres.
Providing an advance warning to the owners of a missile attack is
yet another applications of a military satellite. The Integrated Missile
Early Warning Satellite (IMEWS)launched first in 1970 has means of
discriminating between real missiles and spurious effects. IMEWS
series satellites are capable of detecting an ICBM within seconds of
ignition. By mid-80s, a network of four satellites, one over Russia, two
over Pacific and Atlantic, and the fourth over South America, had been
established.
Meteorological forecast has a special significance for military
planners. The military satellites for the purpose provide advance infor-
mation to the mission planners of reconnaissance satellites on whether
the target area will be clear of cloud cover or not when the reconnais-
sance satellite makes its plan, thus eliminating wastage of limited
reconnaissance film. DMSPIBlock 5D series satellites provide weather
information to both military and civilian areas.
Detecting nuclear explosions and reporting its yield and other
characteristics is another function being performed by some of the early
warning satellites such as IMEWS. The idea of nuclear detection using
satellites originated in the United States and is now being done on a
routine basis.

Fig. 10.27
536 MICROWAVES AND RADAR

Problem 10.7 :Refer to Fig. 10.27. The basis o f a satellite orbitting

around earth is the centripetal force ( F l )due to earth'sgravitation acting
towards the center of the earth balancing the centrifugal force (Fz)acting
away from the center. Calculate the centrifugal force for a satellite of
mass 100 kg orbitting with a velocity of 8 k m l s at a height of 200 k m
above the surface of earth. Assume mean radius of earth to be 6370 km.

Solution :Centripetal force

where m = mass of satellite
V = orbital velocity
R = mean radius of earth
H = height of satellite above surface of earth
The centripetal force balances the centrifugal force.
mv2
Therefore, Centrifugal force = -
(R+ H )

- 100 x (8000)~
(6370+ 200)x lo3
- 64 x 10'
6570 x lo3
= 974 Newtons

Problem 10.8 : A satellite i n a n elliptical orbit has an apogee of

30,000 k m and a perigee of 100d km. Determine the semi-major axis of
the elliptical orbit.
Apogee + Perigee
Solution : Semi-major axis =
2
- 30000+ 1000
2

Problem 10.9 :The farthest and the closest points i n a satellite's

elliptical eccentric orbit from earth's surface are 30,000 k m and 200 k m
respectively. Determine the apogee, the perigee and the orbit eccentricity.
Assume radius of earth to be 6370 km.
Solution : Apogee = 30000 + 6370 = 36370 k m
-
Perigee =200+6370=657Okm
Apogee - Perigee
Eccentricity =
2a
where a = semi-major axis of the elliptical orbit
Apogee + Perigee
Also, a =
2
SATELLITES AND SATELLITE COMMUNICATION 537

or 2a = Apogee + Perigee
Apogee - Perigee
Therefore, orbit eccentricity =
Apogee + Perigee

Problem 10.10: Refer to Fig. 10.28 showing a satellite moving i n

a n elliptical, eccentric orbit. Determine the apogee and perigee distances
if the orbit eccentricity is 0.5.

Fig. 10.28
Solution :The distance from center of ellipse (0) to the centre of
earth (c) is given by ( a x e ) where ( a )is the semi-major axis and ( e )is
the eccentricity.
Therefore, a x e = 14000
a=--14000 - 28000 km
0.5
Now apogee = a(l +e)
= 28000 (1 + 0.5)
= 42000 km
Perigee =a (1- e)
= 28000 (1- 0.5)
= 28000 x 0.5
= 14000 km

Problem 10.11 :Determine the escape velocity for a n object to be

launched from surface of earth from apoint where earth's radius is 6360
km. (G = 6.67 x lo-'' ~ m ~ / kand? M = 5.98 x loz4 kg)
Solution : Escape velocity =
Here r = 6360 k m = 6360 x lo3 m
= GM = 6.67 x lo-'' x 5.98 x
538 MICROWAVES AND RADAR

= 39.8 x 1013 ~ t n ~ / k ~
Therefore,
Escape Velocity = d[2 x 39.8 x 1 0 ~ ~ ] / [ 6 3 6x0 lo3]
= d(79.6 x 101°)/6360

Problem 10.12 :Calculate the orbital period of a satellite in an

eccentric elliptical orbit shown i n Fig. 10.29.
Solution :Semi-major axis, a = -
50000 - 25000 k m
2
Orbital time period, T = 2rr

Fig. 10.29
p = GM = 6.67 x 10-l1 x 5.98 x loz4

= 39250 seconds
= 10 hours 54 minutes

Problem 10.13 :A satellite moving in a highly eccentric Molniya

orbit having the farthest and the closestpoints as 35000 k m and 500 k m
respectively from the surface of the earth. Determine the orbital time
period and the velocity at the apogee and perigee points (Assume earth's
radius = 6360 km).
Solution : Apogee distance = 35000 + 6360 = 41360 k m
Perigee distance = 500 + 6360
SATELLITES AND SATELLITE COMMUNICATION 539

= 6860 km
Semi-major axis of elliptical orbit, a =
41360 + 6860
2

Orbital time period, T = 2rc 4a3/p

= 10 hrs 20 minutes
Velocity at any point on the orbit is given by:
V = dp [ ( 2 / r )- ( l / a ) ]
V = GM = 6.67 x lo-'' x 5.98 x = 39.8 x 1013 ~ m ~ / k ~
At Apogee point, r = 41360 k m

At perigee point, r = 6860 km

Therefore, u = 4 3 9 . 8 x 1013
2
6860 x 1
lo3 - 24110 x lo3

Problem 10.14 : The sum of apogee and perigee distances of a

certain elliptical satellite orbit is 50000 km and the difference of apogee
and perigee distances is 30000 kms. Determine the target eccentricity (e).
Solution : If (ra)and (rp)are the apogee and perigee distances,
then
540 MICROWAVES AND RADAR

Problem 10.15: The semi-major axis and the semi-minor axis of

a n elliptical satellite orbit are 20,000 k m and 16000 k m respectively.
Determine the apogee and perigee distances.
Solution: If (ra)and (rp)are apogee and perigee distances respec-
tively,

then, semi-major axis - ra + rp

2
Semi-minor axis = 6

Therefore, ra + rp = 40000 k m
= 16000
Therefore, ra rp = 256000000
Now ra + rp = 40000
ra x rp = 256000000
Substituting the value of (rp)from (2) in (1)

= 32000 km, 16000 km

ra = 32000 km as it cannot be 16000 km if the semi - major axis is
20,000 lun.

Problem 10.16 : A geosynchronous satellite moving i n a n

equatorial circular orbit at a height of 35800 k m above the surface of
earth gets inclined at a n angle of 2 ' due to some reasons. Calculate the
maximum deviation i n latitude and also the maximum deviation i n
longitude (with reference to longitude ofascendingnode).Also determine
maximum displacements i n kms caused by latitude and longitude
displacements. (Assume earth's radius = 6364 km).
Solution : Height of orbit = 35800 krn
Earth's radius = 6364 km
SATELLITES AND SATELLITE COMMUNICATION 541

Therefore, Orbit radius, r = 35800 + 6364

= 42164 km
Angle of inclination = 2"
Maximum latitude deviation from equator due to inclination (i)is
given by :
. o
hmax=t=2
Maximum l o n ~ t u d deviation
e from ascending node, vmax is given
by :

= 0.0175"
Maximum displacement (in km) due to hmax is given by :

Dx (max) = a i - where ( i )is in deg

= 1471 km
Maximum dispIacement (D.+I)due t o (ymax)
is given by

Problem 10.17: Ageosynchronous satellite orbitting at 42164 km

porn earth's center has a circular equatorial orbit. The orbitgets inclined
due to some reason and it is observed that the maximum displacement
due to latitude deviation is 500 km. Determine angle of inclination (i)
between the new orbital plane and the equatorial plane.
Solution : Maximum displacement Dn (max) due to latitude
deviation is given by:

where L a x = maximum latitude deviation

=z (angle of inclination)

i = Dx (max)/r = -500 - 0.012 rad

42164
= 0.68 deg
542 MICROWAVES AND RADAR

Problem 10.18 :Ageostationary satellite mouing in an equatorial

circular orbit is at a height of35786 km from earth's surface. Zfthe earth's
radius is taken as 6378 km determine the theoretical maximum coverage
angle. ~ l s determine
o the maximum slant range.
Solution : For theoretical marrinlum coverage angle, eleva-
tion - E = 0. Maximum coverage angle,
aygle,
=2 sin' [A
Re+H cos E )
where ' Re = ~arth's radius
H = Height of satellite above earth's surface
' 6378
2 am, = 2 sin-' cos 0o
[k786 + 6378
= 17.4"
If (D)is the maximum slant range, then

' sin [.+ " (

sin-' Re + I1 rns E)]

+ - 2 x 6378 x 42164 x sin 8.7"

= (637812 (42164)'
=40678884+1777802896-537843984~0.1512
=I737139041
D = 41679 km
SATELLITES AND SATELLITE COMMUNICATION 543

Review Questions
1. Differentiate between Geosynchronous, Geostationary and
Sun synchronous orbits. Why are sun synchronous satellite
orbits suitable foi remote sensing satellites ?
2. Distinguish between Spin Stabilised and Three Axis Body
Stabilised satellite. Give their relative merits and demerits.
3. What do you understand by Earth Segment and Space Seg-
ment of a satellite communication link ? Briefly describe the
components of these segments.
4. Briefly describe the payload that you would encounter
aboard a
(a) Communication satellite
(b) Meteorological satellite
( c ) Remote sensing satellite.
5. What do you understand by (a)Point to point connectivity and
(b) Single point to multipoint connectivity ? Where do these
configurations find application ?
6 . Write short notes on :
(a) Satellite TV broadcast
(b) Satellite for weather forecasting
(c) Satellite Navigation
(d) Earth observation satellites.
7. Briefly describe the Molniya series of satellites. Discuss their
relevance for use as communication satellites for the countries
in the polar region.
8. With the help of a diagram, detine the following parameters.
(a) Orbit inclination
(b) Orbit eccentricity
( c ) Apogee
(d) Perigee
(e) Semi major axis

Problems
1. A satellite is orbitting in a near earth circular orbit at a
distance of 640 km. Determine its orbital period (Assume
R = 6360 Km)
[lhour 37 minutes]
2. For an eccentric elliptical satellite orbit with an apogee and
perigee points at a distance of 50,000 lun and 8000 km respec-
tively from the center of earth, determine semi-major axis,
semi-minor axis and orbit eccentricity.
C29000 km,20000 km,0.7241
544 MICROWAVES AND RADAR

3. Ageostationary satellite moving in an equatorial circular orbit

is at a height of 35786 km from earth's surface. If the earth's
radius is taken as 6378 km, determine the Maximum coverage
angle if the minimum possible elevation angle is 5".
[17.32O]
4. An earth station is located at 30" W longitude and 60"N
latitude. Determine the earth station azimuth and elevation
angles with respect to a geostationary satellite located at
50" W longitude. (Assume orbital radius = 42164 km and
earth's radius = 6378 km).
1157.2" , 19.8 "1
5. A satellite is in a circular equatorial orbit at an altitude of
10,000 k m from earth's surface. Determine the maximum
eclipse time is a day during the full eclipse period. (Assume
earth's radius = 6378 km).
[44.5 minutes]

SELF EVALUATION EXERCISE

I Multichoice Questions
One of the following laws governs the movement of artificial
satellites in earth orbits.
(a)Newtonian laws of mechanics
(b)Laws of quantum mechanics
( c ) Galilean laws
( d )Kepler's laws
If (G)is the earth's gravitational constant and (M) the mass of
earth, then for a circular orbit of radius (R), the satellite's
orbital velocity (V) is given by :
(a)V= (b)V= dmiizz

One of the following communication satellites is in a highly

eccentric, inclined orbit.
( a )Molniya series ( b )naduga
(c) Ekran ( d )Gorizont
With reference to satellite communication systems, doppler
frequency shift is absent in one of the following satellite
systems
(a) Geostationary satellite
(b) Near earth orbit satellites
(c) Domestic satellite systems
SATELLITES AND SATELLITE COMMUNICATION 545

( d ) Geosynchronous satellites
5. One of the following expressions can be used to compute the
eccentricity ( e ) of an eccentric elliptical orbit with apogeeand
perigee distances of (ra)'and (rp)respectively.
rp
(a)e = - (b) = (n)
ra + rp
ra

(c)e = 1-(
ra - rp
-
(dl e =
ra
rp
6. Satellite transponders
(a) use a single frequency for reception and retransmission
of information to and from earth
( b ) use alower frequency for reception and a higher frequency
for retransmission
( c ) use a higher frequencyfor reception and a lowezfrequency
for retransmission
( d ) none of these
7. A geosynchronous orbit is the one when the
(a) satellite is placed in an orbit 15,000 miles above earth's
surface in a north to south orbital path
(b) the satellite is placed 22,000miles above earth's surface
in an orbit that matches one of the longitudinal lines of
earth
( c ) the satellite is placed 35,000 miles above earth's surface
in equatorial orbit
(d) the satellite is placed 22,000 miles above earth's surface
in an orbit in equatorial plane
(e) the satellite's orbital velocity is in synchronism with the
earth's rotation
8. A spin stabilised satellite
(a) uses solar panels whose cells are continually oriented
towards the sun
(b) uses solar cells mounted on a cylindrical body that is
continually rotated so that about 40 percent of the cells
receive solar radiation at a given time
( c ) uses gyroscopic action of a spinning satellite
( d ) both (a)and (b)
( e ) both (b)and ( c )
(f, both (a)and (c)
9. A body stabilised satellite
(a) uses solar panels whose cells are continually oriented
towards the sun
546 MICROWAVES AND RADAR

(b) uses solar cells mounted on a cylindrical body that con-

tinually rotates so that about 40 percent of the cells
receive solar radiation at a given time
(c) uses a spinning satellite to,maintain orientation in space
(d) uses a spinning momentum wheel to maintain orientation
is space
(e) both (a) and (c)
0 both (b) and (dl
@) both (a) and (d)
10. The location of a geostationary satellite is always given in
terms of
(a)a certain longitude (b) a certain latitude
(c) longitude and latitude (d) distance from earth's surface
11. The orbital velocity of the satellite
(a) is directly proportional to its distance from earth's surface
(b) is inversely proportional to square root of its distancefrom
earth's center
(c) depends upon the thrust with which it is launched
(d) is continuously changing as the satellite revolves
12. The first orbital velocity for launching a satellite at a given
height above the surface of earth turns out to be 6 kmls. If the
said satellite is launched with a horizontal velocity of 8 kmls,
the satellite orbit would be
(a) circular and eccentric (b) elliptical and eccentric
(c) circular (d) elliptical
13. The multiple satellite access technique suitable only for digital
transmissions is the
(a) Time Division Multiple Access (TDMA)
(b) Frequency Division Multiple Access (FDMA)
(c) Code Division Multiple Access (CDMA)
(d) Both (a) and (b)
14. The commonly used multiple access technique that suffers
from the disadvantage of likelihood of intermodulationdistor-
tion is
(a) Time Division Multiple Access (TDMA)
(b) Frequency Division Multiple Access (FDMA)
(c) Code Division Multiple Access (CDMA)
(d) Packet Access
SATELLITES AND SATELLITE COMMUNICATION 547

15. The minimum number of geostationary satellites needed for

uninterrupted global coverage is

16. With reference to satellite orbit, Apogee is the

( a ) highest point in the orbit
( b ) lowest point in the orbit
( c ) point in the parking orbit
( d ) name given to the boost motor that puts the satellite in
the right parking slot
17. With reference to satellite orbit, Perigee is the
( a ) point in an intermediate orbit
( b ) highest point in the orbit
( c ) lowest point in the orbit
( d ) none of these
18. The satellite series from ISRO (India) meant for earth obser-
vation applications is the
( a )SROSS-C series ( b )INSAT-2 series
( c ) IRS-series ( d )INSAT-1 series
19. The satellite series developed by ISRO (India)for carrying out
scientific studies is the
(a)SROSS-C satellite series ( b )IRS series
( c ) INSAT-2 series ( d )LANDSAT series
20. When a satellite of mass ( m )is orbiting earth at a height ( h )
from the surface of earth with a velocity (u), then the
centrifugal force acting on the satellite away from the center
of the earth having a mean radius of ( R )is given by
( a )mv2/h ( b )r n v 2 / ( +~ h)
(c) m v / ( R + h) ( d )m v 2 / ~
21. For an eccentric elliptical orbit with eccentricity (e)and semi-
major axis ( a ) ,the apogee point distance from the center of
earth having a radius of (R)is given by
( a )ra = a ( 1 + e ) ( b )ra = a ( 1 - e )
( c ) ra = a ( 1 + e ) / R ( d )ra = a R / ( 1 + e)
22. If the apogee distance from earth's center is eight times the
perigee distance from earth's center, the orbit eccentricity is
approximately
548 MICROWAVES AND RADAR

(a)0.12 (i) 0 . i 8

(c)0.4 (d)0.6
23. When the orbit eccentricity (e) equals zero, the orbit is
(a)a parabola (b)a hyperbola
(c)elliptical (d)circular
24. The orbit of a geostationary satellite with an orbital velocity
of approximately 3 kmls is slightlyinclined at 2' to equatorial
plane. The magnitude of the velocity impulse required to be
given perpendicular to orbital plane to correct the orbit in-
clination is equal to (Given that sin 2" = 0.035, cos 2" = 0.999)
(a)105 mfs ( b )3 W s
indeterminate
(c) (d) none of these
25. Mark the wrong statement
(a) an inclined polar orbit can never be a geosynchronous
orbit
(b) All geostationary orbits are geosynchronous orbits
whereas all geosynchronous orbits are not geostationary
orbits
(c) Molniya orbit is popular with Russian communication
satellites
(d) A spin stabilised satellite uses a simple technique for
stabilisation or attitude control but large number of solar
cells whereas a body stabilised satellite uses a very com-
plex stabilisation mechanism but a fewer solar cells.
26. If (a)is the semi-major axis of elliptical transfer orbit and (r)
is the distance of a given point on the orbit from the centre of
earth, then the orbital velocity of the satellite at that point is
given by (p= GM)

27. The eccentricity of an elliptical orbit with the apogee point at

43,000 krn and the perigee point at 7000 k m from centre of
earth is
(a)0.72 (b) 0.16
(c)0.45 (d)indeterminate from given data
SATELLITES AND SATELLITE COMMUNICATION 549

28. The satellite in which the antennas are mounted on a de-spun

platform is the
(a) geostationary satellite
(b) sun synchronous satellite
(c) spin stabilised satellite
(d) 3-axis body stabilised satellite
29. Satellite (S1)and (S2)are orbiting in two different equatorial
circular orbits. The radius of S1-orbit is four times the radius
of S2-orbit. The orbital period of S-1orbit will therefore be
(a) 4 times the orbital period of satellite S2
(b) 8 times the orbital period of satellite S2
(c) twice the orbital period of satellite S2
(d) same as the orbital period of satellite S2
30. With reference to the solar eclipses caused to a communication
satellite by the earth and the moon,
(a) the eclipse due to earth is of a smaller duration
(b) the eclipse due to earth affects the satellite much more
than the eclipse due to moon
(c) while the eclipse due to earth is harmful, the one due to
moon is useful
(d) while the eclipse due to earth is useful, the one due to
moon is harmful
11. Incomplete Statements
1. "An artificial satellite at the correct distance from earth would
make one revolution every 24 hours and thus would remain
stationary above the same spot." This statement was made by
a famous British science fiction writer in 1945 in the journal
Wireless World. His name was
(Arthur C. Clarke, Johannes Kepler, Issac Newton)
2. "The line joining the center of earth and the satellite sweeps
over equal areas in equal intervals of time." This is one of the
famous
(Kepler's laws, Newtonian laws, Galilean laws)
3. The earth observation satellites are placed in or-
bits.
(sun synchronous, geosynchronous, geostationary)
4. In a satellite communication link, if the allowable minimum
angle of elevation of the earth station antenna increases, the
satellite's coverage area .(increasestoo, decreases,
remains unchanged)
5. A larger diameter dish antenna, operational frequency
remaining same, produces a beamwidth.
(narrower, wider)
550 MICROWAVES AND RADAR

6. If the perigee and apogee distances from earth's center in an

elliptical eccentric orbit are both changed by factor (K),the
orbit eccentricity would
(remain unaffected, change by the same factor K, change by a
factor 2 K)
7. In an elliptical satellite orbit, the center of earth lies .
(at one of foci of ellipse, at a point on its semi-minor axis)
8. The multiple satellite access system that suffers fiompresence
of intermodulation products is
(FDMA, TDMA)
111. Identify TRUE a n d FALSE Statements
1. All geostationary satellites are geosynchronous while the
reverse is not true.
2. The velocity of the satellite at a certain point in the orbit is the
same for elliptical and circular orbits as long as the point in
question is at the same distance from center of earth.
3. The velocity of the satellite at its apogee point is always
greater than the velocity at its perigee point.
4. Molniya orbit is the Russian name for a geostationary orbit.
5. The communication satellites with extendible solar sails are
three-axis stabilised.
6. The idea of geostationary satellite was first expressed by
Johannas Kepler.
7. A 6 GHd4 GHz satellite communication link needs a C-band
transponder.
8. The up-link frequency is always higher than the down link
frequency.
SATELLITES AND SATELLITE COMMUNICATION 551

Glossary of satellite related terms

and definitions
Apogee :The point in the terrestrial orbit which is the farthest from
earth.
Ariane space : A private limited company established for the
purpose of producing, financing and marketing European Space
Agency's (ESA)launch vehicle.
Artificial gravity : Use of centrifugal force to simulate weight
reaction in a condition of free fall. It may be achieved by spinning the
vehicle to make the centrifugal force of outer periphery, or bodies within
the vehicle to replace the weight reaction experienced at the surface of
earth.
ASAT :Anti satellite.
Astronautics: The science and technology of space flight.
Attitude : Orientation of a space vehicle as determined by the
relationship between its axes and some reference plane. The system
that achieves and maintains the required attitude.
Azimuth :Bearing or direction in the horizontal plane.
Blackout (radio) : A temporary loss of ratio communication be-
tween a spacecraft re-entering the atmosphere and the ground stations
due to an ionised sheath of plasma that develops around the vehicle.
Booster :The first stage of missile or a rocket.
Centripetal force :A force that is directed away from the center of
rotation.
Centrifugal force :A force that is directed towards the center of
rotation.
Countdown :Acount in inverse numerical order in hours, minutes
and finally in seconds, of the time remaining before the launch of a
rocket.
Doppler effect :The apparent change in frequency due to relative
motion between the source and the observer.
Drag :The resistance offered by a gas or liquid to a body moving
through it.
Equatorial orbit : An orbit in the plane of equator.
Escape velocity : The precise velocity necessary to escape. The
velocity necessary to escape from earth's surface, for instance, is 11.2
Ms.
Exhaust velocity :The velocity of the exhaust leaving the nozzle of
a rocket motor.
552 MICROWAVES AND RADAR

Free fall :The motion of any unpowered body moving in a gravita-

tional field.
g :acceleration due to gravity.
Gesdesy :The science of earth's shape.
Geocrationary orbir :A circular orbit in which a satellite moves
from west to east at such a velocity so as to r e m k fixed to an observer
on earth's $ d a c e . Also known as Clarke's orbit.
g - f g r c ~:A force caused by acceleration expressed in g's.
GTO: Geostationary transfer orbit.
r .tcii;zation :The angle between an orbit path and the equator.
inerCia1 guidance : An on-board system for launch vehicles and
spaceera;%where the gyros, accelerometers and other devices satisfy
guidancerequirements.
L u n c h complex : The complex of site, facilities and equipment
used to launch a missile or a space rocket.
Launchpad: The load bearing base from where a rocket or a space
craft positioned on its launcher is fired.
Launch Windows :m e time interval during which a space vehicle
can be launched to accomplish a given mission.
Lip off or its launch pad blast off: The start of a rocket's flight
from its launch pad.
Mach Number :The ratio of the speed of a vehicle (or of a liquid or
gas) to the speed of sound.
Magnetic storm : A disturbance of the earth's magnetic field
initiated by a solar flare cr sun spot.
Magnetosphere :The region of space surrounding earth which is
dominated by magnetic fidd.
Mass ratio :Ratio of the total mas8 of a rocket vehicle to the mass
that is left when all the propellant is consumed.
Multistage Rocket :-4 racket having more than one stages operat-
ing in succession, each being discarded as its job is done.
Orbit :The path of a body acted upon by the force of gravity. The
word orbit is usually associated with the continuous path of a body
which does not impact with its primary.
Orbital period :Time period for completing one orbit.
Orbital velocity :The velocity necessary to overcome the gravita-
tional attraction of the earth so as to keep a satellite in orbit.
Parking Orbit : Orbit in which a space vehicle awaits the next
phase of its planned mission.
SATEUITES AND SATELLITE COMMUNICATION 553

Payload :Useful cargo of a space vehicle. A rocket's payload may

be a communications satellite and a satellite's payload may be a
transponder.
Perigee :The point in a terrestrial orbit that is nearest to the earth.
Pitch : Movement of a spacecraft about an axis which is perpen-
dicular to its longitudinal axis. It is the degree of elevation or depres-
sion.
Polar orbit :An orbit which passes over the poles.
Pnaary :The body around which a satellite orbits.
Propellant : A solid or liquid substance burnt in rocket for the
purpose of producing thrust.
Radio telescopes :A receiving station for detecting radio waves
emitted by celestial bodies or space vehicles.
Thrust :A propulsive force measured in lb-kg or Newtons.
Trajectory : The flight path of a projectile, missile, rocket or
satellite.
Transponder: The equipment that receives a signal, processes and
then retransmits it at a differeat frequency.
Umbilical Connector : A flexible connector which maintains
ground supplies to a rocket before launch.
Yaw :The rotation of a vehicle about its vertical axis.
zero 'g' :A state when the gravitational attraction is opposed by
equal and oppositeinertial forces and the body experiences no mechani-
cal stress.
 Electronic Warfare
lectronic Warfare (abbreviated as EW) is a high technology

E weapons discipline that is essential or more appropriately the

key to successin modern warfare. It has evolved from the efforts
made all over to optimally utilise the electromagnetic spectrum
to safeguard one's both strategic and tactical security interests and
defeat those of one's enemies. In essence, how best a side can exploit
the electromagnetic spectrum to its advantage is the basis of all EW
concepts and techniques. The whole evolution of electronicwarfare and
its subsequent expansion to the dimension it has acquired today rests
on the philosophy of denying to the opposition an efficient use of
electromagnetic spectrum without affecting one's own exploitation of
the same. The EW discipline therefore encompasses in itself all those
systems and techniques that achieve this objective. A side that is strong
in EW technology can win the battle and even the war. For a side that
ignores EW and does not use it, the results could be one of those of a
disaster. What happened during the Gulfwar in 1991 provides an ample
evidence of this when the sophisticated Iraqi radars were blinded and
their command/communication set up was thrown out of gear by the
countermeasures employed by Allied forces thus giving them total
control of air space.
This chapter gives an overview of the different components of
electronicwarfare and the developments that have taken place in each
of these areas. The chapter begins with an introduction to electronic
warfare to be followed by detailed account of passive and active
electronic warfare techniques. The chapter concludes with a brief on
the future trends in electronic warfare in foreseeable future.
The important topics covered in this chapter include:
11.1. Electronic Ware- 11.6. chaff .and Decoys
fare 11.7. Electronic Optic
11.2. EW Techniques Countermeasures
and Systems (EOCM)
11.3. Electronic Sup- 11.8.Electronic
port Measures Counter Counter-
11.4. Use of Stealth measures (ECCM)
Technology 11.9 Current and Fu-
11.5.E l e c t r o n i c ture Trends in
Countermeasures EW.
(ECM)
ELECTRONIC WARFARE 555

11.1. What is Electronic Warfare !

As stated above, electronic warfare refers to weapons or warfare
discipline in which one side uses equipment and techniques to deny to
the opposition use of their systems of defence and attack on land, in the
air and at sea without compromising the full use of their own systems
meant for the same purpose. The methods for instance used by an attack
aircraft to temporarily disable or confuse enemy's radars to avoid
detection which would enable it to penetrate deep into enemy territory
without making its intentions known would be classified under the
heading of electronic warfare. Similarly, the equipment used on land,
in the air and at sea to provide information on the capabilities of enemy's
systems such as their communication systems, radars, weapons etc.
would also be categorised as the EW equipment. Knowledge of type of
sensor or weapon, its location and information like a fire control radar
is locked onto the aircraft or a missile has been launched towards the
target allows air crew to avoid the radar cover or to take evasive action.
Another example of the EW is use of techniques to make your
command/communication systems, radars, missiles etc. immune to the
tactics employed by the enemy's forces to. confuse, disable or deceive
your systems. For instance, if your radar is so designed that it continues
to do its jobinspite of passive and active techniques employedby a target
aircraft, use of such techniques would also fall in the category of EW.
~lectronic warfare therefore spans from electronicreconnaissance used
to gether information on enemy's command and communication set-up,
radars and other sensors to use of advanced designs for these sensors
to make them completelyimmune to these reconnaissance methods and
information gathering systems and further to the use of passive and
active techniques to avoid identification and detection by enemy's
sensors.
11.2. Techniques and Systems
Different Categories
In order to meet the primary objective of electronic warfare tech-
niques and systems to deny to the enemy the eEective use of electro-
magnetic spectrum and to protect the friendly systems using
electromagnetic spectrum against an EW attack, there are three basic
categories into which EW techniques and equipment can be grouped
into. These are :
1. Electronic Support Measures (ESM)
2. Electronic Countermeasures (ECM)
3. Electronic Counter Countermeasures (ECM)
Electronic Support Measures
Electronic Support Measures (ESM) refer to that division of
electronic warfare which involves actions taken to search for, intercept,
locate and identify radiated electromagnetic energy for the purpose of
immediate threat recognition. Different types of intercept receivers
such as Radar Warning Receivers (RWRs), Laser Warning Receivers
556 MICROWAVES AND RADAR

(LWRs) and Electronic reconnaissance systems can be considered as

the electronic support measures. Though the electronic support
measures provide a source of informahon for immediate action involv-
ing electronic countermeasures and counter countermeasures, evasive
actions, targeting etc. and mainly comprise of various types of intercept
receivers including RWRs, LWRs etc., the systems of electronic recon-
naissance such as Communications Intelligence (COMINT),Electronic
Intelligence(ELINT)are also considered to be belonging to the category
of electronic support measures.
Electronic Countermeasures (ECM)
Having detected, identified and located the threatening sensors
like radars or weapons like missiles, the Electronic Countermeasures
do the job of degrading the performance which could be reduction in the
detection range of the radar or deceiving the guided missile away from
its intended path and so on. Electronic countermeasures use active
techniques to achieve this. It could be some kind of jamming technique
like noise jamming for radars whereas weapons could be made ineffec-
tive by using some kind of deception technique or deploying decoys.
Fundamentally, effective deployment of electronic countermeariures
buys time for the user to evade enemy's defences and succeed in its
mission.
Electronic Counter Countermeasures (ECCM)
While the basic purpose of using electronic countermeasures is do
deny the radar detection of target or deceive it to follow a wrong target
or an electronically created nonexistent target, the purpose of ECCM
techniques is to defeat this and make the systems like radars, missiles
and other similar systems immune to countermeasures. ECM and
ECCM are intimately connected. Information on the countermeasures
adopted by a side enables the other side to employ effective counter
countermeasures against it. Similarly, information on the type and
capabilities of radars and weapon systems used by a side enables the
other side to use effective countermeasures.
11.3. Electronic Support Measures (ESM)
Electronic support measures refer to that branch of electronic
warfare which includes techniques and equipment used for searching,
intercepting and identifying the type and location of radiated
electromagneticenergy for the purpose of immediate threat recognition.
Electronic reconnaissance which means gathering information on the
technical characteristics,deployment mode and operationalcapabilities
of both the potential allies as well as potential enemies is a vital
component of Electronic Support Measures. The other equipment that
is included in the category of ESM, as outlined above, are mainly the
different types of warning receivers such as radar warning receivers,
laser warning receivers, missile approach warning receivers and so on.
ELECTRONIC WARFARE 557

Electronic Reconnaissance
Success in electronic warfare to a great extent depends upon
comprehensive knowledge and understanding of various aspects of the
electronic environment that is likely to prevail in the battlefield
scenario. Electronic reconnaissance ie defined as the detection, iden-
tification, evaluation and location of foreign electromagnetic radiation
emanating from sources other than radioactive phenomenon or nuclear
detonations. Electronic reconnaissance is further classified as :
(a)Signal Intelligence (SIGINT)
( b )Radiation Intelligence (RINT)
(c) Telemetry Intelligence (TELINT)

Radiation Intelligence (RINT) is basically the intelligence derived

from collection and analysis of non-information bearing electromag-
netic radiation emitted unintentionally by foreign equipment, devices
etc. other than those generated by detonation of nuclear weapons.
Telemetry Intelligence (TELINT) is the intelligence derive from the
interception and analysis of foreign telemetry. Signal Intelligence is
further sub-divided into Electronic Intelligence (ELINT) and Com-
munications Intelligence (COMINT).While ELINT is the intelligence
information derived from non-communications type electromagnetic
radiation emanating from foreign sources other than radioactive sour-
ces or nuclear detonations, COMINT is the intelligence information
derived from communications type electromagnetic radiation emanat-
ing from foreign sources by those other than the intended recipients.
The SIGINTsystem is ~suallyan integrated set-up of high perfor-
mance antennas, receivers, signal processing equipment, wide band
recorders and so on. The SIGINT equipment should not be confused
with the equipment used as other electronic support measures like
radar and laser warning receivers because the two types widely differ
in purpose. While the ESM equipment serves the main purpose of
providing information that would lead t o an immediate action in the
form of initiating electronic countermeasures or countercounter-
measures or even in the form of tactical deployment of forces, SIGINT
equipment basically provides both communications and non-com-
munications information for a long term strategic use includinggenera-
tion of electronic data bases to support tactical operations.
The SIGINT system can be installed on a variety of ground based,
airborne or sea platforms. The location and type of platform and the
nature of SIGINT equipment depends upon the location and technical
characteristics such as emission frequency of the target emitters and
also the geographical and political constraints faced by the side gather-
ing information. If the target emitter was radiating around HF range,
the information could be collected by the SIGINT station from a distance
far more than the radio horizon. If the target emitter used HF sky wave
propagation, the signal from the emitter could be intercepted from a
distance of thousands of kilometers. For targets emitting in higher
558 MICROWAVES AND RADAR

frequency, interception and location is limited to line-of-sightdistances.

Airborne and space borne platforms extend the radio horizon still
further. Many developed countries have satellite based SIGINT collec-
tion capability. Satellites can overfly the intended enemy territory
during peace time as well as war time to gather intelligence. Function-
ally, the SIGINT equipment on the satellite will be similar to that
ELECTRONICWARFARE 559

carried in an aircraft. Since the satellite orbits can be predicted to a

very good accuracy, SIGINT can be denied to the interceptor by switch-
ing off the transmitter whilst the satellite is above the horizon.
Fig. 11.1shows the photograph of a long range airborne SIGINT
system capable of providing strategic, operational and tactical informa-
tion on hostile forces by means of its ELINT (Electronic Intelligence)
and COMINT (Communication Intelligence)components.
ESM and Radar Warning Receivers
While electronic reconnaissance and surveillance involve a long
term observation and characterisation of a potential opposition, a
warning system in general is meant for the purpose of detecting,
identifying and locating a threat approaching the system to be
protected. All warning scenarios have a protected entity which could be
an aircraft, a ship, a ground based armoured vehicle etc., an immediate
danger such as illumination by a radar emission or a laser beam or an
approaching missile and an environment containing a variety of begin
objects that must be distinguished from the potential threat. Based on
the nature of threat, common warning receivers or systems include the
Radar Warning Receivers (RWRs), Laser Warning Receivers (LWRs)
and Missile Approach Warning Receivers (MWRs).
Radar Warning Receivers (RWRs) and other ESM receivers are
capable of detecting emissions, mainly radar emissions, over a wide
band and are therefore different from conventional radar receivers
which are designed to receive the well defined signal waveforms used
by the particular radar system. The RWRs not only intercept emissions
over a wide band, they do it without losing any of the characteristics of
each emission that is likely to be used for identification process. The
frequency range of an ESM receiver is typically from 0.5 to 18 GHz and
that of an RWR from 2 to 18 GHz. Efforts are on to extend the frequency
range to 40 GHz in near future.
The parameters on which RWR and ESM receivers can derive
information from the intercepted signal include the RF, the pulse width,
polarisation, PRR, Direction and time of amval, Frequency agility etc.
If the emission can be intercepted for a period more than fraction of a
second,it can reveal a lot of information on the scanningcharacteristics
of the emitter. As an example, if the emitter was a fire control radar,
one could determine the time instant of lock-on which is vital for
initiating countermeasuresor taking other evasive action. Information
on the above mentioned parameters also allows various emitters to be
classified into categories such as Search, Fire control, Navigation
radars and so on.
There is a large variety of receivers available for the purpose of
ESM and radar warning. The performance of different types ofreceivers
is compared on the basis of characteristics like Noise factor, Receiver
560 MICROWAVES AND RADAR
sensitivity, Measurement accuracy, Bandwidth, Dynamic range, Prob-
ability of intercept, False alarm rate (FAR), Throughout and so on.
One of the simplest and most basic types of receivers is the crystal
video receiver. In its basic form, it i; ii~herentlyextremely wide band
and with a suitable antenna pattern, it has the potential of providing
nearly 100% probability of intercept for received signals exceeding
threshold. Fig. 11.2 shows the block schematic arrangement. The
diagram is self explanatory. The law cost, high probability of intercept
and possibility offindinginformation on directionof arrival (DOA)make
this type an attractive option for radar warning applications.

RF Video
Amp. Amp.

Filter Wer
Det .

Tuning

Fig. 11.2

Video
Amp.
IF
Filter Band Pass
Filter
t

Tuning

Fig. 11.3
The superhet configuration is the other receiver type used for the
purpose. The basic block schematic is shown in Fig. 11.3. The tunable
RF filter helps to eliminate spurious outputs caused by intermodulation
products generated in the mixer. The diagram is otherwise self ex-
planatory. Superhet receiver configuration is widely used with various
direction finding techniques particularly where angle of arrival (AOA)
is derived from phase information. This is due to the reason that
frequency and phase information is preserved in a superhet receiver
during conversion of IF and can therefore he measured accurately. The
ELECTRONIC WARFARE 561

high selectivity of superhet nxeiwx leads to only a small portion of

electromagnetic spectrum being sampled by the receiver with the result
that overall probability of intercept is low. However, the relatively
narrow bandwidth gives the best combination of sensitivity and
dynamic range. High sensitivity allows target emitters at far-offranges
to be intercepted through their side lobe and back lobe radiations.
562 MICROWAVES AND RADAR
A wider total bandwidth can be achn3ved in scanning superhet
receivers where the recriver respectivelj scans In frequency across the
desired bandwidth. Such a receiver can be made very effective against
long pulse and CW signals by integrating the output of the video
detector. If a prior information on the emission frequency of the poten-
tial target is available from other intelligence sources, the scanning
superhet receiver can be programmed for optimum search of a portion
of the spectrumcorresponding to specific emitters rather than sweeping
the entire band.
Another approach to increasing the overall bandwidth of a super-
het type receiver is to use a set of fixed frequency filters at the input
before the mixer. The outputs of the filters can be switched to the mixer
followed by IF amplifierfilter and an appropriate detector. Fig. 11.4
shows the block schematic arrangement of such a receiver type. Yet
another approach is to simultaneously convert the outputs of filters to
a common IF. This effectively produces a set of parallel superhet
receivers. This receiver possesses the sensitivity, the dynamic range
and the frequency resolution of a simple superhet receiver and at the
same time has capability to process signals across a very wide RF
spectrum, typically 2 t o 18 GHz. The probability of intercept is nearly
100%.
The channelised receiver described above is usually very complex
and expensive. The same result could be achieved by sweeping a single
receiver across the same band in a time period equal to or less than the
shortest pulse to be intercepted. In this type of receiver called Micros-
can/Compressive receiver, the local oscillator is swept in frequency
linearly across a bandwidth equal to sum of RF and IF bandwidths of
the receiver. The mixer output is fed to a dispersive delay line (DDL)
which produces at its output a compressed pulse at a time measured
from the local oscillator frequency sweep. Multiple input RF signals
appear as a series of narrow pulses at the output (Fig. 11.5). These
receivers have excellent sensitivity and probability of intercept ap-
proaching 100%.They are particularly useful for intercepting pulse to
pulse frequency agile emitters.

Mixer
Filter Outpu!

Local
t

Fig. 11.5
Acousto-optic receiver is another type of receiver used for the
purpose. The operation of an acousto-optic receiver is based on the
deflection of a laser beam as i t interacts with acoustic wave propagating
ELECTRONIC WARFARE 563

in a piezoelectric material such as Lithium Niobate. This device is also

known as Bragg cell. The acoustic wave is excited on the piezoelectric
substrate by the RF signal with acoustic frequency depending upon RF
frequency. The acoustic wave causes periodic variation in refractive
index of the material as it propagates. It is this refractive index
variation that causes the laser beam to deflect. The deflected laser light
is then made to fall on an array of photodiodes.
A Bragg cell receiver consists of a laser source with the laser beam
passing through the piezoelectric crystal which is acoustically excited
by an RF signal. The laser beam is scattered through an angle that is
proportional to the ratio of the wavelength of laser to the wavelength
of source in the crystal. The scattered light is collected on the diode
array which is read out electronically.
Simultaneous signals on different frequencies will each produce
its own deflected beam at an angle proportional to its frequency. It is
effectively a channelised receiver and the total bandwidth of an ESM
receiver based on this technology can be extended by tuning the front
end before it is down converted to Bragg cell operating frequency.
Fig. 11.6 shows the photograph of a typical Radar Warning
Receiver. This particular system (Type RWS-100)is a broadband radar
warning receiver covering a frequency range of 2-18 GHz with 360"DF
(Direction Findig) Capability.

Fig. 11.6
Laser Warning Receivers
Laser Warning Receivers (LWRs) like RWRs also have to en-
counter two related but inherently different scenarios. One of them is
to use these receivers for general monitoring of the battlefield to look
for various types of laser systems operational in the battlefield, their
characteristics, location etc. for subsequent use to formulate battle
strategies. In this type of situation, the information provided by the
564 MICROWAVES AND RADAR

LWRs is not used immediately. In the other scenario which is tactical

in nature, the purpose is to protect the platform on which the LWR is
mounted. The platform could be an aircraft, a helicopter, a ship, a
satellite or even a ground vehicle. An LWR in such a situation provides
information on the impendingthreat, which could be the platform being
illuminated by a laser designator / rangefinder so as to allow the
platform take an evasive action or initiate a countermeasure, an electro-
optic countermeasure in this case.
Fig. 11.7 shows the photograph of a Laser Warning Receiver (Type
LWS-200).The system is capable of providing warning of laser threats
from rangefinders, target illuminators, target trackers and missile
guidance systems.

Fig. 11.7
LWRs have assumed great significance in the modern warfare
environment due to rapid poliferation of different types of lasers used
in a variety of military applications. Larger the number of lasers in use
in the battlefield, greater is the need to equip oneself with ways and
means that can offer protection against them. The lasers that are a
usual sight in the modern warfare include Nd-Glass and Nd-YAG lasers
used as rangefinders and designators, semiconductor lasers used in
short range rangefinding and communications and beam rider applica-
tions, CO2 lasers also in short range rangefinding, laser radar and beam
rider applications and so on. Eye safe lasers in 1-3 pm band are fast
replacing Nd-YAG lasers. Solidstate vibronic lasers like Alexandrite
ELECTRONIC WARFARE 565

and Titanium Saphirre lasers which offer tunable output wavelengths

in visible and near infrared to reduce the countermeasurevulnerability
of fixed wavelength YAG lasers. Then there are lasers emitting 3-5
pm band as countermeasures for heat seeking missiles, 8-12 pm lasers
for FLIR countermeasures. Weapon lasers is another category and the
lasers belonging to this group mainly include high energy solidstate
lasers, high power CO2 lasers, chemical and excimer lasers and free
electron lasers. Such a wide variety of lasers covering a wavelength
range from ultraviolet to infrared put very stringent requirements on
LWRs. An LWR, to be effective, must respond to this wide band, a
situation similar to the one encountered in case of RWRs. The other
important considerationsare the high probability of detection, low False
Alarm Rate (FAR) and a high sensitivity.
The laser parameters that can be of importance or in other words
that LWR should determine to achieve the desired objective depend
upon the application. For LWRs that serve only to alert the platform
and define the threat, what would usually be sufficient is to make coarse
measurement of laser wavelength, pulse duration and pulse repetition
rate (PRR).The information is considered adequate for distinguishing
between weapon grade lasers, designatorslrangefinders, communica-
tion lasers and lasers for countermeasures. Weapon lasers emit in
specific wavelengths and in long duration pulses, rangefinders are
Q-switched lasers operating at a low PRR, designators are similar to
rangefinders but have relatively much higher PRR. Countermeasure
lasers are also similar to rangefinders but of high energy. Communica-
tion lasers are either modulated CW lasers or high PRR pulsed ones.
On the other hand, for LWRs that are directly linked to counter-
measure lasers, it would be required to measure some more charac-
teristics in addition to those listed above. Also, the parameters would
be required to be measured with a much greater accuracy. The issue of
threat localisation,that is, determining the location of the threat laser,
is far more complex and demanding in case of LWRs than it is in case
of RWRs. Often, the receiver intercepts energy coming from a direction
other than direction of the source due to scatteringlreflection and the
directional data provided by the LWR could be misleading. If the
receiver intercepts the direct beam, threat localisation is simpler. The
level of accuracy required in finding this information also depends upon
the scenario. In most of the cases, quadrant localisation is adequate
whereas in some airborne LWRs, it may be necessary tp determine the
same to an accuracy of a few degrees.
Another aspect of LWRs is providing warning on strategic target
observables that have prominent EO/IR signatures such as Intercon-
tinental Ballistic Missiles (ICBMs),Submarine Launched Ballistic Mis-
siles (SLBMs), Cruise missiles, strategic aircrafts etc. ICBMs and
SLBMs emit strongly in infrared as a result of the intense IR radiation
from their rocket exhaust plumes. This is due to tremendous amount of
heat energy being released in the rocket exhaust due to large quan 'ty
of fuels being burnt to achieve the desired thrust. The most o en f!
566 MICROWAVES AND RADAR

considered emission bands for ICBMISLBM missile warning are 2.7 and
4.3 pm bands. Bands considered for cruise missiles and strategic
aircrafts are in the 8-12 pm atmospheric window. In these bands, the
EOAR observables result from the air vehicle itself rather than its
exhaust plumes.

Fia. 11.8
Fig. 11.8 shows block schematic arrangement of a simple laser
warningreceiver. It consists ofreceiving optics at the front end to collect
the laser energy followed by optical filters to select the laser
wavelengths of interest. These filters are very narrow band bandpass
filters that pass only the desired wavele~gth.The radiation after
passing through the filter is focused onto the detector or scanned across
an array of detectors. The electrical signals are processed to extract the
information un laser parameters.
Missile Approach Warning Receivers
Missile approach warning is very important as according to a
claim, more than 90 per cent of all military aircraft losses are due to
passive IR guided missilas. Thus there is 2 continuing requirement for
an effective missile approach warning system that is effective against
IR guided missiles and is suitable for mounting on a variety of aircraft.
Detection range is often the key factor determining the choice of system
as this would decide the time availrhle to the platform to be protected
to take an evasive action or initiate countermeasures. In other words,
the detection range is a function of the required minimum warning time,
the speed of the aircraft, speed of zpproaching missile and direction of
attack.
Detecting a passively guided missile from a tactical aircraft plat-
form effectively enough as to allow the air crew take an evasive action
or initiate a countermeasure is a complex job. There are two commonly
used approaches to building missile approach warning receivers
(MWRs).The first uses an active radar system, usually a pulse doppler
or a CW radar, which provides information on the range, velocity and
position of the missile. The other C;ype of MWRs are the passive IR
systems which depend for their operation on their ability to detect some
emission characteristics of a guided missile. There are three primary
emissions that the designer of MWR can exploit. The first is the
radiation from the exhaust during the launch or boost phase of the
missile's flight when the fuel is burnt at the maximum rate and radiant
ELECTRONIC WARFARE 567

intensity is also maximsum. In this case, the receiver operates in the

ultraviolet or 3-5 pm band. The second is the radiation produced from
the combustion products of sustainer motor if missile uses one. For this,
receiver operates in 8-14 pm band having an atmospheric window of
low attenuation as the received intensity in this case is low. These
systems are capable of detecting missiles at larger ranges prior to their
burnout. Active radar systems are overt in nature and suffer from
limited detection range. On the other hand, passive IR systems offq
covertness at the expense of limited information on range and velocity.
One of the approaches to designing an MWR system is to make use of
good points of both active and passive techniques. The overall system
in that case would be a combination of an approach warning radar and
a passive missile approach warning receiver.
Fig. 11.9 shows the photograph of a Missile Approach Warning
System (Type PVS 2000) which is basically a pulse doppler radar
capable of detecting the approach of missiles from rear and initiate
countermeasures automatically. The system is particularly effective
against infra-red guided surface-to-air and air-to-air missiles.

Fig. 11.9
The major problems faced by the designers of MWRs are the same
as the ones faced by the designers of RWRs or LWRs. These are the
need to have a high probability of detection and a low false alarm rate.
A low false alarm rat is very crucial as the countermeasures must be
initiated very rapidly. If the countermeasures involved the use of flares
or other consumables, a high false alarm rate could rapidly exhaust the
supply or expose the presence of the aircraft to the enemy.
568 MICROWAVES AND RADAR

In the earlier paragraphs, we have looked at the passive means of

detecting enemy sensors and weapons, in the paragraphs to follow, we
shall look at the passive means of avoiding enemy sensors like radar.
11.4. Use of Stealth Technology
Stealth technology refers to all those techniques used to minimise
target signatures. In the earlier paragraphs, we have discussed dif-
ferent types of warning receivers used by air crew to warn them of a
possible detection by an enemy radar (Radar Warning Receivers),or an
electro-optical system (LaserWarning Receivers) or of a possible missile
attack (MissileApproach Warning Receivers). These systems assist the
air crew take an evasive action and route their flight away from the
cover of enemy sensors and weapons. Another way of looking at this
problem is to look for ways and means to avoid detection, to be more
practical, minimise the detection range of radar and other sensors. This
will allow the air crew to penetrate deep into enemy's territory and
approach very close to the enemy radar without being detected. All this
can be achieved by rninimising the target signatures which include all
those characteristics of the target, an aircraft for instance, that can be
used as the basis of detection by enemy sensors and weapons. For
example, it would be radar cross-section of the on the target echoing
area in case of a radar sensor, target contrast in addition to the echoing
areain case of an electro-opticalsensor or emissionfrom the hot portions
of the aircraft body or radiation from the combustion products in case
of passive IR guided missiles.
The relevance of using stealth technology- was amply demons-
trated in the recent past during the Gulf war of 1991 when extremely
low radar and IR signatures of F-117A and B-2 aircraft made them
virtually undetectable and play havocon difficult to attack Iraqi targets.
Use of stealth technologies reduces the enemy's reaction time by reduc-
ing the detection range.
Radar cross-section (RCS) is o i e parameter that designer of the
stealth technologies have exploited the most firstly because a reduction
in RCS reduces the detection range and secondly because a reduced
RCS leads to a reduced S/N ratio at the radar receiver input which will
degrade its tracking accuracy. It may also be mentioned here that the
detection range of a radar is proportional to only the fourth root of the
RCS and a 100:lreduction in RCS would yieldonly 3.16 times reduction
in detection range. However, the jamming power or the decoy size
required for an effective electronic countermeasure would be directly
proportional to the RCS. So, reduction in RCS gives an overall benefit.
Reduction of Radar Cross-section
The basic purpose cf reducing radar cross-section is to reduce the
amplitude of the reflected signal. This could be achieved by either
shaping the vehicle in such a way as to minimise reflections or using
some special materials or paints that absorb electromagnetic energy.
When an aircraft is designed with the primary objective of achiev-
ing an optimum aerodynamic performance, which has been the ap-
ELECTRONIC WARFARE
570 MICROWAVES AND RADAR

proach until quite recently, the resulting aircraft is more likely to have
physical features that would act as strong reflectors of electromagnetic
energy and thus make the aircraft highly vulnerable as it approaches
the enemy sensors. These features include skin joints, panel edges,
plane surfaces, re-entrant cavities of the structure and systems under
dielectric panels. Reflections from internal and external surfaces,
angles and sharp edges of air intakes can get reinforced by those from
complex surfaces of engine's compressor, stator blades, flat plate radar
antennas and bulkheads behind nose radomes.
There are two approaches to i-eshaping the aircraft. The first is to
reduce the RCS to a minimum by eliminating at the design stage as
many as possible of all those features that are strong reflectors of
electromagnetic energy. The amount of energy reflected from such
features depends upon their orientation to the incident wave and upon
the relationship between the phases of individual reflected waves.
Resonant conditions need to be avoided as far as possible by eliminating
small components.
In another approach, the aircraft is so designed as to be made of
such faces and wing leading and trailing edge angles which reflect radar
energy away from the source. Fl117A and B-3fighter aircraft are good
examples of use of stealth technologies (Fig.11.10).
Radar Absorbant Materials
The other approach is to build the aircraft structure as far as
possible with materials that absorb radar energy. Though there is an
increasing use of composite materials with good RF energy absorbing
ability, use of metallic parts can not be avoided completely. There are
two basic types of non-reflecting materials. The first of these types
depends entirely on the absorbing properties of the material. The second
type depends upon the interference phenomenon to effectively reduce
the amplitude of the energy reflected from the material. The energy
reflected from the air-material surface primary boundary and the other
reflected from the material backing interfere destructively to minimise
reflections. While absorbing materials are usually effective over a wide
frequency range, materials relying on interference are effective over a
limited frequency range.
Some radar absorbant paints have also been developed for the
purpose. These paints can be used to cover most of the surface of the
aircraft and are effective in reducing surface currents and creeping
waves which otherwise would radiate as a component of the reflected
signal.
The aircraft also need to minimise their visible and IR signatures
necessitated by increasing use of IR and other electro-optical sensors.
The main sources of IR radiation are engine components,jet pipes, jet
exhaust and aerodynamic heating in case of high speed aircraft. All
these factors need to be looked into by property shielding the jet pipes,
cooling the hot components,avoidingdirectline of sight with hot turbine
components and so on. Detection in the visible band depends for its
ELECTRONIC WARFARE 571

operation on the contrast between the target and the background. The
contrast can be minimised by using suitably coloured finishes. Another
possible appr~achcould be to equip the aircraft with lights capa.ble of
generating lighting of intensity and colour matching that of the back-
ground thus considerably reducing the visible signatures. Reduction of
acoustic signatures is particularly important in case of helicopters.
11.5. Electronic Countermeasures
Active and Passive Techniques
While different types of warning receivers discussed in the earlier
part of the chapter can be used to detect and identify the enemy's
sensors and weapons to enable the target take an evasive action or
initiate countermeasures, the stealth technologies help the target avoid
the cover of those sensors bv being. undetectable to them until it has
u

gone very close to the sensor. The electronic countermeasures are used
to degrade the 2erformance of enemy's sensors and weapons like
reducing the detection range of a radar or forcing the missiles to deviate
from their intended trajectory and move away from the target. The ECM
basically buys time for the ucer to evade enemy's defences during the
crucial stage of an engagemen; or battle. In the paragraphs to follow,
we shall briefly describe both passive and active countermeasure tech-
niques including different types of jamming, use of chaff and different
types of decoys.
11.6. Jamming Techniques
Jammingrefers to transmitting some kind of signal so as to either
swamp the target return or other wanted signals or to induce range,
velocity or angle errors into the target system. The former is known as
the Noise Jamming and the latter goes by the name of Deception
Jamming. While a noise jammer is usually effective against a much
wider range of threats, a deception jamming system is limited in its
application to a very small group of threats. Deception jammer requires
to transmit much lower jamming powers, is smaller in size but is very
complex to design. On the other hand, noise jammer requires to trans-
mit large powers in the jamming signal, is larger in size and volume.
The noise jammers are sometimes used by the enemy sensors to their
own advantage when they use the jarnrningsignal as a beacon and select
the home-on-jam mode in a guided missile. The two types of jamming
techniques are described in little more detail in the following para-
graphs.
Noise Jamming
As mentioned above, a noise jammer basically transmits a noise-
like signal with its main objective being to maximise the noise power
in the receiver's bandwidth. The noise signal completely masks the
desired signal which could be a communication signal or an echo signal
in case of a radar. The efficacy of noise jamming clearly depends upon
the ratio of the desired signal strength to the jamming signal strength
and also on the nature of the noise signal. The first condition is obvious
572 MICROWAVES AND RADAR

and the second arises from the fact that the noise signal should be such
that it is indistinguishable from the receiver's internally generated
noise once it enters the receiver. It is because of this reason that the
ideal noise waveform is the white gaussian noise, that is, it has a
gaussian amplitude distribution and a uniform spectral density.
Some of the methods used for generating the noise ~i~nal'include
Direct Noise Amplification (DINA),where thermal noise is amplified
over the bandwidth of interest using wideband amplifiers like TWTs,
Frequency modulation of a carrier with wideband noise or a combina-
tion of wideband noise and a sawtooth or a similar waveform, Frequency
modulation of carrier by a narrowband noise, Amplitude modulation by
noise and so on.
The performance of a noise jammer can be explained by writing
the equation for the jamming signal to the desired signal ratio at the
receiver. In case of a radar receiver, this equation can be written from
radar principles as :

where J = Received power due to the jamming signal (undesired sig-

nal)
S = Received power due to the target (desired signal)
PJ = Jammer power
GJ = Jammer antenna gain in the direction of the radar
RJ = Range from jammer to radar
GRJ = Radar antenna gain in the direction of jammer
PT = Radar peak power or equivalent power if radar integrates a
number of pulses
GT = Radar antenna gain in the direction of the target
RT = Range from radar to target
h = Radar signal wavelength
o = Radar cross-section of the target
If the jammer signal bandwidth (BJ) is greater than that of the
receiver (BR),the JIS equation above gets modified to :

Ratio [$)is taken to be equal to unity in case of (BJ) being equal

to or less than (BR).

Self-ProtectionJammer :If the noise jammer is carried on the
aircraft for self protection, as is the case for a self-protectionjammer,
ELECTRONIC WARFARE 573

= RT and
then the aircraft itself is also the radar target. In that c a s e R ~
GRJ = GT and the J I S gets further modified to :

The equation for ( J I S )can be used to estimate the range at which

the radar starts detecting the target or the jammer stops shielding it
from the radar detection. This range called the Burnthrough Range can
be determined for a self protection jammer from :

The derivation of the above equation assumes that the radar

receiver noise is much smaller than the jamming noise and is therefore
ignorable. The above equation also assumes that the jamming signal
has the same polarisation as that of the radar signal. In general, the
jamming effectiveness falls off 2s the self-protection jammer ap-
proaches the radar because the target signal as received by the radar
is varying as ( 1 1 ~whereas
)~ the jamming signal is varying as (11~)'
only.
Stand-Offand Escort Noise Jamming :In case of Stand-Offor
Escort Jamming, the noise jammer is not carried by the aircraft to be
protected. Instead, it is carried by another aircraft flying outside the
range of local air defences and providing the noise jamming power to
protect the attacking aircraft as it flies to and from the target in case
of stand-off jamming. In case of Escort Jamming, the jammer is on-
board a specialised ECM aircraft that escorts the attack aircraft. The
burnthrough range of a stand-off jammer is much greater than that of
a self-protectionjammer.
Spot Jamming :A Spot Jammer is the one that concentrates its
power into a narrow band centered on the frequency of the enemy's
sensor approximately matching its bandwidth. Therefore, for given
effective radiated jammer power, a spot jammer produces maximum
noise power spectral density in the receiver.
Barrage Jamming :A Barrage Jammer is a broad band jammer.
It spreads its radiated energy over a large bandwidth. The noise power
spectral density at the receiver gets diluted in case of barrage jamming.
Swept Spot Jamming : A Swept Spot Jammer provides the
advantage of high noise power spectral density of a spot jammer and
large bandwidth of a barrage jammer. It does so by rapidly sweeping
the spot noise signal across a wide band of frequencies. This jammer
can be made very effective by appropriately choosing the sweep rate
and the bandwidth.
Cross-PolarisedJamming: Cross-PolarisedJammer radiates a
jamming signal whose polarisation is at 90' to the polarisation of the
wave radiated by the radar to be jammed. It is particularly effective
against radars that employ side lobe blanking technique to improve the
574 MICROWAVES AND RADAR

quality of their PPI display. The side lobe blanking ensures that the
radar does not respond to the signals entering via side lobes. It does so
by using an auxiliary antenna channel whose effective gain is lower
than that of the radar antenna main lobe but higher than the radar
antenna side lobes. The radar receiv ?r does not respond to any signal
whose amplitude in the auxiliary antznna channel is greater than that
of the main antenna channel (Fig. 11.11).If the jamming signal has a
cross-polarised component of significant strength, there is every
likelihood of cross-polarised main lobe pattern (shown dotted in Fig.
11.11)falling below the effective gain of auxiliary antenna with the
result that 211 signals in the main lobe could get suppressed while the
side lobe signals wouid 5.. accepted. One way to reduce the sensitivity
of the radar receiver to cross-polarised jamming is to use another
auxiliary channel with a cross-polarised antenna with its gain being
larger than the cross-polarised gain of the main antenna and much less
than its gain for the designed polarisation.

, Radar antenna response

'_-- -.
\Radar an\tenna
!,. .
cross polarised response
\ I J \
Auxilia

I c Angle
L'

Fig. 11.11
Deception Jamming
While a noise jammer attempts to swamp the radar echo signal by
transmitting towards the radar a noise signal of sufficient power over
the bandwidth of the radar receiver, a deception jammer transmits a
corrupted signal towards the radar which it accepts as a genuine one.
A deceptionjammer in most of the cases accepts the radar signal, carries
out a suitable modification and then retransmits the same towards the
radar. The modification usually involves changing an appropriate char-
acteristic of the signal so as to induce deception in range, velocity or
angle errors. In other words, the jammer creates a number of false
targets for the radar to detect.
Fig. 11.12 shows the block schematic arrangement of a basic
deception jammer. The jammer accepts the radar signal which is then
modified as per the deception program. The modified signal is then
amplified before it is retransmitted. A deception jammer is usually a
ELECTRONIC WARFARE

Receiving Antenna
\
RF
Ampl~fier - Signal
Modification

I
-

Tranmittinq Antenna
1 Programme I
output
Amplifier

Fig. 11.12
self protection device, one of the basic requirements for the aircraft. The
output amplifier shown in Fig. 11.12 ensures this. The required total
gain of the system is given by :

Deception i n Range :Range deception is particularly effective

against search and track radars. Thejammer in this case creates alarge
number of false targets at different ranges and angles from the true
target position and thus confuses and overloads the radar operator as
well as radar processing circuits. In order that false target pulses
radiated by the jammer are indistinguishable from the real target echo
pulses, one of the methods is to transmit replicas of radar pulses shortly
after or before the arrival of the radar pulse. If the radar pulses are
stable in terms of PRF, RF, their arrival time can be anticipated and
false target pulses could possibly be transmitted either before or after
arrival of the pulse. But in case the radar employs a varying PRF or
pulse to pulse RF agility, the false target pulses can be sent only after
the receipt of the real radar pulses. The range deception jammer first
receives the radar pulse, amplifies it to make it stronger than the target
skin echo and transmits it back towards the radar without introducing
any change in the range information. This is done to ensure that the
signal is received by the radar range gate alongwith the genuine signal.
The stronger false signal overloads the processing circuitry and the
operator tends to reduce the sensitivity of the receiver. The false signal
this way succeeds in capturing the range gate. Having done that, the
jammer progressively introduces a change in range all the time staying
within the range gate, the phenomenon being known as the Range Gate
Pull-Off (RGPO).
Deception i n Velocity :A velocity deception jammer maximises
jammer energy in radar's Doppler filters. The deception jammer first
576 MICROWAVES AND RADAR

receives the radar signal, amplifies i t and then re-transmits it without

any change in frequency so that i t is received by the Doppler filter
processing the skin echo signal. The jammer signal is kept 5 to 10 dB
stronger so that it captures the filter and the velocity gate. The jammer
then progressively shifts the RF of its signal pulling the target away
from the true target, a phenomenon similar to that of Range Gate
Pull-Off (RGPO).
Angle Deception :Inducing range and velocity errors degrades
the tracking performance of the radar. The tracking errors induced,
however, are very small. If large tracking errors are required to be
induced, angle deception would be needed. The techniques used for
angle deception would depend upon the trackingmethodology employed
by the radar. For instance, technique used for trackingradar employing
sequential lobbing is entirely different fiom the deception technique to
be employed against radar using monopulse tracking. Detailed descrip-
tion of these techniques would be beyond the scope of the present text.
Modern aircraft emply jammers that have the capability to
produce a variety of jamming signals. These jammers could either be
pod mounted or configured inside the aircraft. Airborne noise and
deception jamming system (TypeANIALQ-167(V))is an example. Fig.

Fig. 11.13
ELECTRONIC WARFARE 577

11.13shows a potograph of its pod mounted version. The system can be

configured for manual or fully automatic control of jamming functions.
The Noise Jamming modes available are continuous spot, intermittent
spot, continuous barrage, intermittent barrage, swept amplitude
modulation and swept noise and fast set on spot noise. The deception
jamming modes available are multiple frequency repeater, velocity gate
stealer, repeater swept amplitude modulation, combination of velocity
gate stealer and repeater swept amplitude modulation, chirp gate
stealer and so on.

11.6. Chaff and Decoys

Chaff
Chaff is basically an airborne cloud of light weight reflecting
objects typically consisting of strips of aluminium foil or metal coated
fibers or nylon which produce clutter echoes in a region of space for the
purpose of confusing, screening or otherwise adversely affecting the
opposition sensor system such as radar. Chaff elements are dipoles cut
to a length that is approximately equal to half of the wavelength of the
radar radiation. It is packaged in the form of cartridges carries that can
be easily deployed. Each cartridge a range of dipoles of slightly varying
lengths so as to make the chaff effective over a wide frequency band.
Fig. 11.14 shows photographs of CFD-100 and CFD-200 type airborne
chaff and flare dispensing system. These are self protection systems
providing protection against radar and IR guided weqpons.
The important characteristics of chaff include its Fall rate, Bloom
time, Effective echoing area, Polarisation and Bandwidth.
Fall Rate is the rate at which chaff falls to earth and it depends
upon factors like type of chaff (Aluminiumfoil, Aluminised fiber, Silver
coated nylon etc.), altitude from which it is dispersed and local
meteorological conditions. Typical fall rate in still air is 0.5'm.k for
aluminium foil, 0.6 m/s for silver coated nylon and 0.3 m/s for
alurninised fiber. Bloom time, a s a e name suggests, is the time taken
by the chaff to reach its maximum echoing area or a sufficient echoing
area for a given application. It mainly depends upon the method
employed to disperse the chaff. It is an important feature and the
requirements vary depending upon the application. For instance, chaff
used for self protection must bloom very rapidly, typically in a few
milliseconds. The Echoing area depends upon the size of the chaff and
also the bandwidth it is supposed to cover. A narrow band chaff of a
given size provides a much larger echoing area than a broad band chaff
of the same size. Polarisationis another important feature. As the chaff
cloud is out of the turbulent region, majority of the dipoles acquire a
horizontal attitude. A portion of the chaff cloud that acquires a vertical
orientation tends to fall more rapidly and thus offering a lower echoing
area. The polarisation performance of the chaff must be considered
while deciding the quantity of chaff required for a given application.
Chaff Bandwidth depends upon the variety of dipole lengths packed
678 MICROWAVES AND RADAR

into a dispensor and subsequently forming the chaff cloud. Another very

Fig.11.14
important feature of the chaff cloud is its spectrum. Ideally, the chaff
cloud should produce the same doppler spectrum as the one produced
by the aircraft the chaff is supposed to protect. The chaff spectrum is
usually narrow and depends upon the wind velocity because most of the
time it spends in the air, it drifts with the wind. However, if the chaff
is dispensed into a turbulent atmosphere, for the brief period, it would
have a wide doppler spectrum and this additional spectral width of he
chaff spectrum can succeed in breaking the lock of a tracking radar.
There are other techniques that can be used to overcome the spectrum
limitations of the chaff. A discussion on those techniques is beyond the
scope of the text.
ELECTRONIC WARFARE

Decoys
Decoy is another electronic countermeasure. There are passive as
well as active decoy systems. A decoy is a dummy radar target much
smaller in size than the aircraft it is intending to protect but whose
reflectivity, or in other words the radar cross-section, is made larger
than the actual target by using reflectors or other means. It confuses
the radar operator as well as the radar processing circuits and thus
increases the survivability of the attacking aircraft. Chaff discussed in
the preceding paragraphs can also be considered as an expendable
radar decoy but it has its own limitations, a narrow doppler spectrum
being one of the major ones. Decoy provides solution to some of these
limitations. There are two commonly used modes of deployment of
decoys. We have towed decoys which are towed by the vehicle to be
protected and free falling decoy deployed in a fashion similar to that of
chaff.
The purpose of the towed decoy (which could be passive or active)
is to provide an echo signal that is much stronger than that from the
towing vehicle so as to force the missile tracker or radar to lock on to
the decoy rather than the actual target. It has an obvious advantage
that the radar or the missile seeker can not distinguish the decoy from
the target on the basis of velocity discrimination. As long as the two,
towing vehicle and decoy, remain unresolved by the sensor (radar,
missile seeker etc.), the sensor will either track the signal center of
gravity ofvehicle-decoysystem or lock on to the stronger source of signal
return i.e. the decoy. In case of aircraft towed decoys, the separation
between the aircraft and the decoy is a few hundred meters. This

Fig. 11.15
 MICRCWAVES AND RADAR
countermeasure is particularly effecti-e g gain st semi-actively guided
missiles. However, the efficacy of the syatem strongly depends upon the
engagement geometry. Fig. 11.15 is the photograph of a towed decoy
(Ariel towed radar decoy). It is an example of off-board electronic
countermeasures to protect aircraft from raGar directed weapons. This
system is said to provide very effective electronic countermeasures to
monopulse radar systems. The technology of Ariel enables it produce
the angle deception essential to defeating this typ of radar design.
Miniature active expendable decoys are also being increasingly
used. This has been made possible by rapid advances in the field of
Monolithic Microwave Integrated Circuits (MMICs).Essentially, these
decoys are miniaturised transponder jammers comparable in size to
chaff dispensers or cartridges, Once deployed, they receive, modify and
re-transmit an amplified version of the radar signal to simulate
semblance of a real target that they are trying to protect. In fact, they
are miniaturised version of deception jammers. When fired from chaff
dispensers, they fall free. They can be fittedto arocket and fired forward
from the aircraft. These decoys have very stringent design require-
ments particularly due to their small size and mode ofdeployment.They
should be capable of withstanding the launch forces, must have a long
storage life once integrated with their storage batteries and so on.
11.7. Electro-opticCountermeasures
Electro-optic countermeasures are employed against systems
operating in the optical spectrum from ultraviolet to infrared like Laser
designators and Range finders, Laser Radar, other electro-optic sensors
and so on. The basic EW concepts discussed earlier with reference to
RF spectrum such as noise and deception jamming, use of warning
receivers, need to minimise target signatures and use of stealth tech-
nologies, deployment of decoys, use of chaff and obscurants are equally
valid when we talk about electro-opticcountermeasures.
Laser warning receivers have been discussed in the earlier part of
the chapter. An effective countermeasure against IR homing missiles
is the use of IR jammers. These jammers transmit spurious modulated
IR energy towards the target so that what the missile receives is the
target return as well as the spurious signal. This degrades missile's
tracking performance. The jammer radiation could also be used to
saturate the seeker front end detector with a little larger power or even
damage the front end optics or detector or IR dome with a high energy
laser. An active IR jammer of this type should radiate energy in the
spectral bandwidth of the seeker to provide adequate jamming to signal
ratio at the seeker. The radiation also needs to be modulated with an
optimum waveform.
, Use of obscurants is a very effective electro-optic countermeasure.
Smoke can be very effective at optical wavelengths. The obscurant can
be scattering type or absorptive or both. It is less effective at IR. It is
not an attractive proposition for airborne deployment due to the prob-
lem for aircraft to deploy smoke in the right place, at the optimum time
ELECTRONIC WARFARE 581

and in large quantities and the problem of wind drift. IR flares and IR
decoys are also very common.
11.8. Electronic Counter-Counter
measures (ECCM)
The purpose of using support measures such as warning receivers
on a vehicle to be protected, like an aircraft, was to detect, identify and
locate enemy's sensor system such as a radar and subsequently follow
it up with an appropriate action which could be an evasive manoeuvre
or initiate a countermeasure. The purpose of using countermeasures
was to make these sensor systems ineffective or degrade their perfor-
mance during the brief period of engagement or attack. Even a momen-
tary unlocking of a fire control radar or degradation in tracking
performance of the radar could give sufficient time to the crew of the
attacking aircraft to fulfill their intended mission.
While the main job of the countermeasure is to minimise the
efficacy of the enemy's sensor systems, the counter-countermeasures
are a counter to the countermeasures. The purpose of the counter-
countermeasures is to equip the systems like radar and other com-

Fig. 11.16
582 MICROWAVES AND RADAR
munication features to defeat the objectives of countermeasures. The
primary objective of ECCM techniques, for instance, when applied to a
radar system is to design the radar in such a way that it accomplishes
its intended task of fire control, tracking, guidance, search etc. depend-
ing upon the type of radar even in the presence of countermeasures
employed by the attacking aircraft. A radar that is designed and built
to avoid saturation of radar receiver front end, maxirnise signal to
jamming ratio, reject false targets created by deception jammers, dis-
criminate against directional interference etc. is said to be equipped
with ECCM features and have a survivability in the battlefield.
Hyghes' low level air defence radar (Fig. 11.16) is one example of
a radar having in-built ECCM features. The said radar performs ery
effectively in severe clutter and ECM environments mainly due to its
low side lobe pencil beams and frequency agile phase beam scanning.
As an another example, a laser designator used on an aerial
platform to illuminate the ground targets and then guide weapons along
the scattered radiation employ some form of coding in the transmitted
pulses that is known to the seeker subsystem. This is done to defeat the
opposition design of directing the weapon such as missile away from its
intended course towards the real target to some other false target if it
chooses to do so by illuminating a false target with'a laser of the same
type.
When it comes to building ECCM features into a radar system, one
needs to take a note of all the subsystems comprising the radar system
nainely the antenna, the transmitter, the receiver and the signal
processor. In the following paragraphs, we shall bridy discuss ECCM
techniques as related to various subsystems of a radar.
Antenna Related ECCM Features
The antenna parameters that can be exploited for building ECCM
features into this part of the radar system include Gain, Directional
pattern, Antenna scan rate and Polarisation. Some types of deception
jammers work on anticipation of the scan rate. Their purpose can be
defeated by using random scanning. Also a high gain antenna can be
used to spotlight a target and burn through the jammer. An antenna
with multiple scans allows rejection of beam containing the jammer
signal and still maintain detection capabilitywith the other beams. Low
side lobe level is desirable in most of the cases except when the
consequent increase in the main beam width worsens the problem of
main beam jamming. Other techniques used to prevent jamming from
entering through side lobes are the Side Lobe Blanking (SLB)and Side
Lobe Cancelling (SLC). The purpose of SLB system is to prevent the
detection of strong targets and interference pulses through side lobes
and is discussed in an earlier paragraph. The purpose of SLC is also
similar in nature and is to suppress noise interference received through
side lobes. This is achieved by estimating the direction of arrival of the
jamming signal and jamming signal power with the help of an array of
auxiliary antennas and then modifying the receiver directional pattern
to place nulls in the jammer direction.
ELECTRONIC WARFARE 583

The polarisation characteristics of antenna can be effectively ex-

ploited to build ECCM capabilities either by keeping antenna cross-
polarisation pattern as low as possible to prevent jamming through
cross-polarisation pattern or by making use of cross-polarisation pat-
tern and co-polarisation pattern and then discriminating the useful
signal from undesired signal from chaff or jammer on the basis of
polarisation.
Transmitter Related ECCM Features
When it comes to building ECCM capabilities into the transmitter
portion of the radar, the parameters normally exploited include Trans-
mitted power, Emission wavelength and the Waveform. Increase in
transmitted power coupled with burnthrough mode can be used to
defeat noise jamming. Burnthrough mode however is not effective
against countermeasures like chaff, decoys, repeaters etc.
Frequency agility and Frequency diversity are very effective
methods of ECCM. A frequency agile radar transmitter usually has an
ability to change its emission frequency on pulse-to-pulse or batch-to-
batch basis. Pulse-to-pulse frequency agility is however not compatible
with doppler processing. The same, however, is possible if the frequency
agility is on batch-to-batch basis. Frequency diversity refers to use of
more than one complementary radar emissions at different frequencies
either from a single radar or from several radars. The objective of using
a frequency agile radar or resorting to frequency diversity is to force
the jammer to spread its energy over the entire bandwidth covered by
the frequency agile radar with the result that jammer output power
spectral density reduces. This makes the ECM less effective.
Narrowing antenna's beamwidth is also a very effective counter-
countermeasure against noise jamming countermeasure as it restricts
the sector that would be blanked by main beam jamming. In addition,
it also provides a strobe in the direction of the jammer which along with
another two or three spatially separated radars would have the same
effect as that of using a larger antenna.
Radar Receiver and Signal Processing Related ECCM Fea-
tures
The other important aspect is to make the radar receiver and
signal processing chain immune to the ECM such as relatively high
power noise jamming signals, signals from deception jammers, reflec-
tion from chaff etc. Jamming signals that are able to penetrate the
resistance offered by the antenna's ECCM abilities, if large enough, can
saturate the radar processing chain which can lead to near complete
elimination of information from the targets. Solution for avoiding such
a situation lies in the use of a wide dynamic range receiver. Two
commonly used approaches for implementing such a receiver are the
Log and Linear-Log receivers.
A Log Receiver is the one whose video output is proportional to log
of the R F input signal envelope over a specified range. It allows the
radar receiver to detect target returns that are larger than the toise
684 MICROWAVES AND RADAR

jamming signal. Log receiver has certain disadvantages. One of them

is that it is only effective against high level noise jamming signals and
allows a low level noise jammer to be more effective by amplifying low
level jamming signals more than high level target returns. The other
disadvantage is that log characteristics widen the spectrum of received
echoes wh'ch makes the life difficult for radars like MTI and Pulse
d o p p k radars.
The s11c:tcomings of log receiver are largely overcome in a Linear-
Logrecetv~rin which the output signal is nearly proportional to the RF
signai envelope f3r high input signal amplitudes and directly propor-
tional to the 2nvelope amplitude for low signal amplitudes.
11.9. Clwrent and Future Trends in
Electronic Warfare
On oxla hand, there have been further developments in the existing
EW conctpts and systems like development of ultra wide band radar,
mult~staticradar,advances in the field of decoys and so on, on the other
hand, some new concepts have also emerged. The definition of
Electronic Warfare (EW)has been gradually undergoing a change over
the last one decade or so. Earlier definition of EW did not include the
use of Anti-Radiation Missiles (ARM) and Directed Energy Weapons
(DEW)as a part of Electronic Warfare. DEW make use of high energy
laser radiation, high power RF energy, prinlarily mcrowaves; or par-
ticle beams with the objective of placing very high energy density on
the target so as to damage it in such a way as to defeat its mission
intentions. It could be blinding a seeker, dazzling the crew or cause
direct destruction through thermal or mechanical shock. The Anti-
Radiation Missile (ARM)having its own antenna, receiver and proces-
sor has the objective of homing on to the victim radar and destroying it
completely. It gets its initial cues from the ESM system and then
continues to home on to the \lctim radar emission through side lobes
or flash of energy in the main beam. The concept of EW has been
recently expanded to not only include ARM and DEW but also the other
weapons of physical destruztion of equipment, facilities and personnel.
The other concept that has emerged fast and assumed significance
is the Information Warfare (IW). The main objective of the doctrine of
Information Warfare is attainment oftotal information superiority over
the opposition. The main reasons for the rapid development of this
concept are ever increasing dependence of nations and their armed
forces on information technology and also the fact that all information
processors and systems are theoretically vulnerable to corruption or
falsification of information, disruption, destruction and manipulation.
Information technology is viewed in military spheres as a major force
multiplier due to its capability to transfer and manipulate huge quan-
tities of data. A false or biased data subtly introduced into potential
adversary's decision making process can, for instance, seriously mislead
them on our intentions and thus influence their response.
ELECTRONIC WARFARE 686

There has also been increasing use of Knowledge Based Systems

(KBS) and Artificial Neural Networks. An area in which a knowledge
based system is already being used is in the radar ESM particularly in
de-interleavin;: several interleaved radar emissions received by the
radar warning receiver. This technology can also be applied for iden-
tification and prioritisation of threats and assisting in making an
optimum choice and timing of countermeasure. Neural networks are
particularly attractive in solving a wide range of signal processing
problems in the fields of pattern recognition, image processing, radar
signal extraction and so on. Artificial neural networks together with
knowledge based system capabilities are expected to provide the best
possible response to the EW environment in a given situation.

Review Questions
1. What do you understand by the term 'Electronic Warfare'?
Differentiate between Electronic Support Measures (ESM),
Electronic Countermeasures (ECM) and Electronic Counter
Countermeasures (ECCM).
2. What is the basic p r p o s e of using Electronic s i p p o r t
Measures (ESM) ? Briefly describe how Radar Warning
Receivers (RWR) constitute a part of an ESM set-up.
3. What do you mean by the term 'Stealth Technology'? How does
this help in providing an effective countermeasure ?
4. Differentiate between passive and active electronic counter-
measures. Briefly compare Noise Jamming and Deception
Jamming type of active ECM.
5. What type of jammers, you think, would be suitable for in-
capacitating a monopulse type of tracking radar ? Briefly
describe the operation of such a jammer.
6. Write short notes on the following :
(a) Decoy
( b ) Chaff
(c) Electro-optic Countermeasures
(d) Directed Energy weapons (DEW)
( e ) ELINT and COMINT
7. What are Electronic Counter Countermeasures (ECCM) ?
Briefly describe the typical features of a modern radar
equipped with ECCM capability.
SELF EVALUATION EXERCISE
Multichoice Questions
1. One of the following is not an ESM system
( a ) Radar warning receiver
( b ) Laser warning receiver
 MICROWAVES AND RADAR

( c ) Communications intelligence (COMINT)

(d) Decoys
Use of decoys
(a) is a passive countpmeasure
(b) is an active countermeasure
(c) could be a passive or an active countermeasure
(d) is a countermeasure deployed from a ground vehicle
Chaff belongs to the category of
(a) Electronic Support Measures (ESM)
(b) Passive Electronic Countermeasures
(c) Active Electronic Countermeasures
(d) Electronic Counter-Countermeasures(ECCM)
Of the following, the most effectivejamming technique against
a tracking radar would be
(a) Noise jamming
(b) Range deception
(c) Velocity deception
(d) Angle deception
For a given quantity of chaff material, the one that is likely to
have a larger echoing area would
(a) also have a wide band response spectrum
( b ) have a narrower band response spectrum
(c) fall very rapidly
(d) be randomly polarised
Side Lobe Canceling (SLB) and Side Lobe Blanking (SLBjare
(a) ECM techniques
(b) ECCM techniques
(c) characteristics of all radar antennas
(d) two different names for the same technique
Mark the false statement
(a) One would always like to have a radar antenna that has
ideally zero side lobe level
(b) Use of radar signal absorbant materials and paints min-
irnise target signatures
(c) For maximum effect for a given jammer power, the jam-
mer output spectrum should be within the radar receiver
bandwidth
(d) Coherent MTI and Pulse compression radar have a good
amount of ECCM capability
ELECTRONIC WARFARE 587

8. The missile that homes onto radiation from radar and has its
own antenna, receiver and processor with the objective of
destroying the victim radar is
(a) Anti-Radiation Missile (ARM)
(b) Ballistic missile
(c) Cruise missile
(d) Air-to-Surface missile
9. Frequency agility and Frequency diversity are
(a) passive countermeasures
(b) active countermeasures
(c) ECCM techniques
(d) radar receiver characteristics
10. Polarisation diversity is
(a) a passive countermeasure
(b) an ECCM technique ,
(c) an active countermeasure
(d) radar receiver characteristic
 Microwave Communication
Link Basic Design
Considerations
Introduction
In the earlier chapters, we have discussed fundamental concepts
related to MICROWAVES in general and Microwave Components in
particular. Microwave devices, both Microwave tubes a s well a s
Microwave semiconductor devices, were discussed a t length. Other
topics that were in focus included Microwave measurements and In-
strumentation. Microwave r a d a r , a device t h a t makes u s e of
microwaves to determine parameters of distant objects is one of the
most significant applications of microwave frequency band. Radar was
discussed in great detail. Satellite communication that relies on
microwave frequency bands for connectivity between two users of the
link was also covered in detail.
The present chapter focuses on the important issues related to a
microwave communication link.
12.1. Microwave Link Design
- Basic Considerations
The important areas that need to be attended to while designicg a
microwave communication link include the following :
1.Path Clearance
2. Path Profiles
3. Field Survey of the path
4. Propagation Aspects
Path Clearance
For all practical purposes, we can assume that microwaves travel
as line of sight signals. This necessitates that the microwave link desig-
ner takes into account the curvature of the earth's surface between two
adjacent microwave stationb while calculating heights of the antennas.
Refer to Fig. 12.1.IfDl and D2 are the distances from the path location
in question to either end of the microwave hop and (HIis the earth's
clprvature as shown in Fig. 12.1,then (H) can be computed from aqua-
fion 12.1.
MICROWAVE COMMUNICATION LINK BASIC DESIGN 589

Where (HI is in feet and Dl, D 2are in miles. (K) is a constant whose
magnitude depends upon the actual propagation of microwave energy
along the path length. For straight line microwave propagation,
K = 1.Different values of 'K' are used to describe wave propagation that
does not exactly follow a straight line path.

'/
Fig. 12.1.
As evident from the above figure, the effect of earth's curvature on
path clearance is the greatest at the center of the path.
Effect qf Atmospheric Density : As the microwave frequency
increases, its behaviour vis-a-vis its propagation through the atmos-
phere gets more and more close to that of visible light. Microwave ener-
gy can be reflected, refracted and focused by atmosphere like light. As
we go higher above the surface of earth, atmospheric density decreases
and as a result, lower part of the signal wavefront tends to travel a
little slower than the upper part of the wavefront. A 'K' value of 4/3 can
be used to describe such a path (Fig. 12.2).Thus the effect of dekreasing
atmospheric density with height is the bending of the signal path inthe
same way as curvature of earth. In such axase, the earth appears flat-
ter than normal and 'K' greater than '1' results in less lower height on
each end of the path.
Earth Bulge : This effect is particularly seen in humid coastal
areas where,an atmospheric condition like that of an inversion layer
can be observed. Due to this condition, the atmospheric density is ob-

Fig.12.2.
590 MICROWAVES AND RADAR

served to increase with height in a certain region. This causes the

microwave signal to bend opposite to the curvature of earth as shown
in Fig. 12.3. The bending may be severe enough to cause the microwave
signal diverted back to the earth. The term 'Earth Bulging' arises from
the curvature of the earth blocking the propagation of the signal. The
value of 'li:factor used to take care of this effect is typically 213.

Fig. 12.3.
Ducting or Super-Refraction : Ducting, also known as Super-
refraction, occurs in a region where the variation in refractive index
with height is far more rapid than it usually is in a normal or standard
atmosphere. Reduction of refractive index with height is usually linear
and gradual. Rapid variation may be caused by certain atmospheric
conditions where there is a layer of warm air trapped over a relatively
cooler air leading to formati n of what is known as Temperature Znver-
9
qipn Rkgion. In this region, the temperature increases with height in-
stead of decreasing at a rate of 6.!j°C/hh in the standard atmosphere.
This usually owprs &ar the surface! bf 'eafth,.typically within 30
meters of it. "?his rapid reduction in refractive index causes a rapid
bending of microwaves towards the earth's surface. The waves travel
within the duct getting repeatedly refracted from the top of the duct
and reflected from the earth's surface. Ducting is illustrated in Fig.
12.4. Duct propagation is used at microwaves to cover distances beyond
the line-of-sight (LOS).

Fig. 12.4.
Different conditions of refraction inside the duct are considered by
MICROWAVE COMMUNICATION LINK BASIC DESIGN

defining a modified refractive index 'N given by & + ( h h ) ]where 'h' is

the height above the surface of earth and 'r' is the true radius of earth.
Another parameter 'M' called Excess Modified Index of Refractive
Modulus is defined as :
M = ( N - 1 ) x lo6 ...(12.2)
A duct is formed in the region where 'Mdecreases with height 'h'
or in other words where the gradient ( d M / d h ) is negative. Figs. 12.5
(a)to (d) depict different Ducting conditions. Fig. 12.5 (a)shows the 'h'
versus 'M curve for a standard atmosphere. Figs. 12.5 ( b ) to ( d ) show
similar curves for Surface Duct, Elevated Duct and Ground Based Duct
respectively.

(c) (dl
Fig. 12.5.
We also define a critical angle 'ad.The waves get trapped inside
the duct only when the angle of incidence or entrance is less than this
critical angle. Also quite expectedly, the ducts behave like waveguides
and are associated with a cut-off frequency or wavelength. Waves
having a wavelength below a certain cut-off frequency can only be
trapped inside the duct. The cut-off wavelength in terms of duct height
is given by the empirical relationship :

where q,,) = wavelength in cm

d = duct height in meters
592 MICROWAVES AND RADAR

De-coupling and Reflective Effects :De-coupling effects cause

the signal at the receiving antenna to reach outside the main lobe of the
receiving antenna's directional pattern. The reflective effects from at-
mospheric sheets cause cancellatior of the main microwave signal.
Effect of Fresnel Zones : If we look a t the cross-section of the
propagating wave front, it can be ousemed to be consisting of con-
centric areas around the basic path line. This is depicted in Fig. 12.6.
Beginning with the area adjacent to the path line, these areas are num-
bered as 1, 2, 3... and so on. Various Fresnel zones are marked in the
diagram shown. The importacce of the Fresnel zones arises from the
fact that blockage cf any part of the wave front or reflection of any
portion of the wave fro:?C from an obstacle on the way affects the signal
that is received at the receiving antenna. If the propagation path is so
chosen that it does not block the area represented by the first Fresnel
zone By any obstructions below the path line, then the path is said to
have first Fresnel zone clearance. The reflection characteristics of the
earth along the path must also be coasidered. There may be partial or
significant cancellation of the desired signal if the clearance equals an
even Fresnel zone number.
The Fresnel zone clearance can oe computed from :

where D = Total path length In Miles

F = Frequency in GHz
Dl, D2 = Distances in Miles from the point in question to each
end of the path
F1= Distance in Feet from the path line to the edge of the
first Fresnel zone
First Fresnel zone
[ rSecond Fresnel zone

Cross-section
of wavefront
Tx
Fig. 12.6.
Clearances for the higher order Fresnel zones can be calculated
from :
FN = F~$T ...(12.5)
MICROWAVE COMMUNICATION LINK BASIC DESIGN 593

The clearance above obstructions changes if the path line gets al-
tered due to any propagation anomalies. There can be variations in the
received signal if the clearance equals that of an even Fresnel number
zone. Things could be worse leading to a deep fading if the surface is
highly reflective.
Path Profiles : I t is important for a microwave link designer to
examine all obstructions along the proposed path line and also the
obstructions to the side of the path line particularly when they are
reflective. Topographic maps are examined for the purpose. There are
several techniques of plotting this information. Some of them are
depicted in Fig. 12.7. Fig. 12.7(a)shows the use of a curved base line to
represent earth's curvature while Fig. 12.7(b)shows the use of a curved
template to draw the path lineover a flat earth base line. Yet another
method is to use a straight earth base line and a straight path line (Fig.
12.7(c))with the necessary clearance over each obstruction calculated
and shown by a symbol. In this case, both the earth's curvature as well
as the desired Fresnel clearance must be calculated and added to plot
the clearance target above each obstruction.
Stra~ght Ray

earth

'Flat earth (b)

' ~ l a tearth
(c)
Fig. 12.7.
594 MICROWAVES AND RADAR

Field Survey of *.hePath :The pach design amve4 a t from the

study of maps needs to be confirmed in most of the cases. A microwave
path designed and installed on the basis of the path profiles drawn
from map study and without the benefit of a field survey is a high-risk
proposition except for short and obvious types of paths. This is usually
done by a field survey which looks a t aspects like potential reflection
points, confirmation of natural obstructions, man made obstructions
and so on. In case of potential reflection points brought out by field
survey, the path can be so designed as to minimise the possibility of
signal cancellation by a signal reflection.
Propagation Aspects : After having determined the physical
characteristics of the microwave path, the next step is to look a t the
microwave signal propagation aspects. Various causes that lead to loss
of signal a s it travelshwards the receiving site are briefly described in
the following paragraphs. All these causes have the combined effect of
reducing the signal strength a t the receiving antenna. The received
signal needs to be stronger than the minimum detectable signal by the
receiver also known a s receiver sensitivity. In fact, the link is so
designed as to oFer a sufficient margin called the Fade Margin, be-
tween the minirqumdetectable signal of the receiver and the actually
received signal. This ensures uninterrupted communication link ser-
vice for all practical purposes. Now, the f i r s t a d the foremost are to
estimate microwave signal attenuation or the path loss.
Attenuation or Path Loss :The path loss can be computed from :
Loss (in dB) = 96.6 + 20 loglo f + 20 logl& ...(12.6)
Where '$ is frequency in GHz and 'D' is path length in miles. Equa-
tion 12.6 assumes that there are isotropic antennas a t each end of the
link. Fig. 12.8 shows a graphical representation of various components
that contribute path loss a s we travel from transmitter output to
receiver input. The first loss element encountered is due to the trans-
mitter waveguide that connects the transmitter output to the transmit-
ting antenna a s depicted in the figure. Transmitting antenna then
provides gain and that is how the signal strength is shown going up-
wards. The line with downward slope represents the free space path
loss. Receiving antenna provides gain followed by some loss in the
receiver waveguide that connects the receiving antenna output to the
receiver input. As we can see, the received signal is shown as being far
more stronger than the minimum detectable signal of the receiver. The
difference between the two is known as the FADE MARGIN. In a typi-
cal case, it may be as large as 40 to 50 dB. The significance of Fade
Margin will become more clear in the following paragraphs.
Multipath Fading : M u l t i p a t h F a d i n g occurs when t h e
microwave energy is scattered or diffracted along the path in such a
manner that a part of the signal arrives a t the receiving antenna out of
MICROWAVE COMMUNICATION LINK BASIC DESIGN 595

t l
Transmitter
I 3
Receiver

ARX .. .
sensltlvity
Fig.12.8.
phase with the desired signal.,The effect is depicted in Fig. 12.9. This
leads to variation in the received signal strength even if the atmos-
pheric conditions are normal and none of the effects like Earth Bulge,
Ducting, Reflections etc. described earlier are present. Multipath
fading is mathematically represented by Rayleigh distribution hnc-
tion. Fig. 12.10 shows the plot of the Probability of fade exceeding the
Fade Margin (in percent) versus Depth of Fade according to the
Rayleigh distribution function. As is clear from the plot, as the permis-
sible fade depth is increased, the probability of cancellation decreases.
This clearly brings out the need for having a greater fade margin. As
an illustration, if the link was designed to have fade margin of only 20
dB, the probability of a fade deep to touch the receiver sensitivity point
and cause the receiver to squelch is a%high as 1%.On the other hand,
for a depth of fade (or the fade margin) of 50 dB, the probability be-
comes as low as 0.001%.

ired S~gnal

/
/-.
',
Scattered S~gnal Scattqred Signal

Fig. 12.9.
596 MICROWAVES AND RADAR

Path Availability (or Reliability) :The probability of Fade Mar-

gin getting exceeded is directly related to the Path Availability or
Reliability. For example, for a probability of 0.001%, it would be
99.999%. Path availability is also expressed in terms of Outage Time
per year, which in this case would be about 5 minutes. This is computed
by multiplying 0.00001 (decimal representation of 0.001%) by total
number of minutes in an year. This is undoubtedly a very good figure
for a single microwave hop. Let us consider a system having 30 tandem
hops. Considering the case where no two hops will fade simultaneously,
a propagation outage time of 150 minutes or 2.5 hours per year may or
may not be acceptable to the designer.
t

Depth of Fade ( d B )
Fig. 12.10.
Propagation unavailability can also be computed from :
~ = [ i l ~ b x 2 . 5 ~ 1 0 -F/lOl ~ ~ ~ ~ 1 0 - ...(12.7)
This expression takes into consideration the frequency of opera-
tion, the path length, the climatic conditions, the topographic factors
and so on. In the above expression :
U = Unavailability Factor in decimal form
It needs to be multiplied by a factor of 5,25,6OO(number of minutes
in an year) to get 'Unavailability' in minutes per year.
A = Topographic Factor
It equals '4' for a very smooth terrain, '1' for an average terrain with
some roughness and '0.25' for mountainous, very rough and dry terrain.
B = Climatic Factor
It equals 0.5 for hot and humid locations such as Gulf Coast, '0.25'
for temperate or northern areas and 0.125 for mountainous or very dry
areas.
'f' is frequency in GHz, 'D' is path length in miles and 'Fis the Fade
Margin in dB. As is obvious from expression 12.7, Mountainous and
very terrain seem to be the best bet as they have least values for factors
MICROWAVE COMMUNICATION LINK BASIC DESIGN 597

'A' and 'B' thus leading to lowest unavailability for a given frequency of
operation, fade margin and path length. This expression does not take
into account effects like Ducting, Earth bulge, Reflections etc. and as-
sumes that the path has been designed for adequate clearance. Also,
this expression is for one way propagation and 'Vvalue should be mul-
tiplied by '2' for a 2-way propagation assuming that outages in both
directions do not occur simultaneously.
Space Diversity :If we look a t the expression 12.7 for the outage
time, it is clear that if a microwave path designer is faced with a situa-
tion of an unacceptable outage time, the options before him to get the
desired outage are operating a t a lower microwave frequency or shor-
tening the path length or increasing the fade margin. However, if
making either of these changes is also unacceptable, then one of the
possible solutions lies in making use of the concept of SPACE DIVER-
SITY. The concept is briefly described in the following paragraph.
Refer to Fig. 12.11(a).In the microwave path shown, there is one trans-
mitting antenna and two receiving antennas. The receiving antennas are
usually spaced 30 to 60 feet in the vertical direction. Antenna spacing be-
tween 30 and 60 feet provides the best compromise between improvement
in reliability of propagation and increase in cost due to increased tower
height. Each receiving antenna has its own dedicated receiver. The base
band outputs from the two receivers are then combined as shown in Fig.
12.11(b).As shown in Fig. 12.11(a),both antennas receive the desired sig-
nal. Since the path lengths for the two reflected signals are different, should
a cancellation occur in one receiving antenna, it is not likely in the other
with the result that probability of a deep fade is reduced.

Receiver
'1'
-
-
Combined
Baseband
Signal
L q --
4 b)
Fig. 12.11.
598 MICROWAVES AND RADAR

The improvement expected by employing space diversity is given

by :
(7x 10-5, <n (s2)(1f/l0)
ISD= D ...(12.8)
where Is3 = Improvement Factor in decimal form
S =Vertical spacing between two antennas in feet
F Frequency of operation in GHz
7

13 = Path length in miles

F = Fade margin (in dB) associated with receiving antenna
with longer transmission line
'!pidl value ofZsD is between 100 and 500.Expressions 12.7 and
2.8 caa be combined to get the expression for 'Unavailability' for the
spsce dh-ersity case. The expression is given in 12.9 below :
a x b x (3.6 x ( D ~(lO-w/lO)
)
UsD = - ...(12.9)
s2
To sum up, Space Diversity is an effective technique of improving
the propagation reliability where multipath or reflective type of fading
is predominant. It is not effective where Ducting, Earth bulge and
similar effects are present.
Problem 12.1 :Amicrowave terrestrial communication link operating
at 6 GHz has a single hop path length of 50 miles. Determine the distance of
thepath line above the surface of the earth mid way between the path length
of this hop fir the case offirst Fresnel zone clemance.Also cakulate the same
fir third Fresnel zone claamnce.
Solution :It is given that :
D = 50 miles, D l= 25 miles, D2= 25 miles and f = 6 GHz
Now, first Fresnel zone distance is given by :

= 72.5 [g] = 104.2feet

Third Fresnel zme distance can now be computed from :

F3 = F 1*= 104.2 x a=
104.2x 1.732= 180.5feet

Problem 126 :The single hop path length of a terrestrial mtcrowave

communication link operating at 4.6 GHz is 40 miles.In order to ensurejust
jirst fresnel zone ctearance, what maximum obstacle height can be tolemted
at a distance of 5 miles f i m the tmrsmitter if the transmitting antenna
height above the surface of earth is 200 feet. Assume a normal K-factor fir
computing earth's curvuture.
MICROWAVE COMMUNICATION LINK BASIC DESIGN 599

Solution :Refer to Fig. 12.12.It is given that :

D = 40 miles, Dl = 5 miles, D 2= 35 miles and f = 4.5 GHz
Substituting these values in the expression for first Fresnel zone
distance, we get :

I:;:[
= 72.2 - = 71.2 feet

In the next step, we shall compute the curvature 'h' of the earth at
a distance of 10 miles from the transmitter. This is given by
[D1.D211.5K ]
For normal case, K = 1 which gives h = 116.6 feet. This implies that
the path line is only 83.4 feet ( = 200 - 116.6) above the surface of earth
at a distance of 5 miles from transmitting antenna. Since the required
Fresnel zone clearance is 71.2 feet, maximum tolerable height of the
obstacle above surface of earth would be'(83.4 - 71.2) = 12.2 feet.
5 miles

735 mi'es 1

Fig. 12.12.

Problem 12.3 :In problem 12.2, if we assume an earth bulge con-

dition and recommend a K-factor of 2 13, will the first Fresnel zone pass
without any obstruction ?
Solution :For K = 2/3. 'h' a t a distance of 5 miles from the trans-
mitter would be :
5 x 35/1.5 (213) = 175 feet which implies t h a t theiavailable
clearance above the surface of earth would be (200 - 175) = 25 feet. The
required first Fresnel zone clearance as computed in the previous prob-
lem is 71.2 feet which means that it would be obstructed.

Problem 12.4 : The 'Unavailability Factor' computed in decimal

form for agiven microwavepath is 2 x Compute the 'Outage Time'
or the 'Unavailability' in hours per year.
Solution :The 'Outage Time' in hours per year is given by :
Outage Time (in Hours per Year)
=2 x low4x 8760 = 1.752 hours per year
600 MICROWAVES AND RADAR

Problem 12.5 :Fig. 12.13. shows the plot of 'Path Unavailability'

versus 'Fade Margin' at a certain known operating frequency and as-
suming the values of topographic factor and climatic factor as 1 and
0.25 respectively. Compute the following from the given plot.
( a )Improvement in the probability of fade margin getting ex-
ceeded for a path length o f 50 miles when the fade margin is
increased from 40 to 50 dB.
( b )Improvement in the probability of fade margin getting ex-
ceeded for a fade margin of 40 d B for a single hop path length
change from 50 to 30 miles.
4 I

I
-
1 I I
30 35 LO 45 50
Fade Marg~n(In dB)
Fig. 12.13.
Solution : ( a ) From the given plot, for a path length of 50 miles
and fade margin of 40 dB, the probability of fade margin getting ex-
ceeded is 7 x 10- 5 . Also the probability of fade margin getting exceeded
for a fade margin of 50 dB is 6 x 10- 6. The probability figure therefore
has improved by a factor of

(b)At 40 dB, for path length of 50 miles, the probability figure has
'.
already been determined in part (a)above to be 7 x 10- For a path length
of 30 miles and fade margin of 40 dB, it can be seen from the plot to be equal
to 2 x 10- giving an improvement by a factor of 3.5.

Problem 12.6 :A given single hop microwave link with a path

length of 40 miles has an 'Unavailability Factor'of 0.01%. What would
be the magnitude of this factor in decimal form if the link had four such
hops. Also determine the 'Unavailability Factor' for the single hop link
if it employed space diversity that led to an 'Improvement Factor'of 100.
Solution :Unavailability Factor (for single hop) = 0.01% = 0.0001
Unavailability Factor (for Chop link) = 4 x 0.0001 = 0.0004
MICROWAVE COMMUNICATION LINK BASIC DESIGN 601

Improvement Factor due to space diversity = 100

Therefore, Unavailability Factor for the space diversity case
= 0.0001/ i O O = l o p 6or 0.0001%.

Problem 12.7 :Determine the 'Outage Time' i n minutes per year o f

a 4-hop microwave link operating at 3.5 GHz and having single hop
path length of 3G miles. Assume a n average terrain with some rough-
ness (a = 1) and a hot and humid climate (b = 0.5). The link is designed
to have a fade margin o f 40 dB.
Solution :The 'Unavailability Factor' for a single hop can be com-
puted from :
~ = a x b x 2 . 5 x 1 0 - ~ x f x ~ ~ x 1 0 - ~ ~ ~ ~
where a = 1, b = 0.5, D = 30 miles, f = 3.5 GHz and F = 40 dB
Substituting these values,
u = ~ x o . ~ x ~ . ~ x ~ o - ~ x ~ . ~ x ~ ~ o o o x ~ o - ~
= 118.125 x
U (in minutes per year)
= 118.125 x 10- x 525600 = 6.2 minutes per year
For the Chop link, U = 6.2 x 4 = 24.8 minutes per year

Problem 12.8 : If the frequency of operation of a single hop

microwave link is increased by 25%, its path length doubled and the
Fade Margin increased by 10 dB, with the topographic and climatic
factors remaining unaltered, what would be the change i n the value of
'Unavailability Factor' ?
Solution : If U 1 and U 2 a r e the values of the 'Unavailability
Factor' in the two cases, then

Now, D ~ = ~ I J1.25f1
~ , a~n =
d F : ! = ~ ~10
+
Substituting these values,

Therefore, the Unavailability Factor remains unaltered.

Problem 12.9 :A microwave link makes use of space diversity to

reduce the 'Outage TimeJ.What would be the improvement i n 'Outage Time'
i f the vertical spacing between the two receiving antennas were doubled
with all other parameters remaining same ?
602 MICROWAVES AND RADAR

Solution :The 'Improvement Factor' is proportional to square of

antenna spacing. Therefore, it will increase by a factor of 4. Conse-
quently, the 'Unavailability Factor' and hence the 'Outage Time' will
also reduce by a factor of 4.
12.2. Microwave Path Design at Millimetric
Wavelengths
Basic considerations for a microwave path design a t operating fre-
quencies of say 20 GHz and higher corresponding to millimeter
wavelengths are no different from what we have already discussed.
Some of the parameters however are likely to be quite different in case
of higher operating frequencies than their values a t lower microwave
frequencies. Another important aspect of microwave links a t mil-
limeter wavelengths is that the path lengths are much shorter, typical-
ly 10 miles or even lower, and it is relatively much easier to establish
line of sight.
Effect of Obstructions and Fresnel Zone Clearance
At lower operating frequencies, paths are relatively much longer,
Fresnel zones are larger and therefore clearances are difficult and
costlier to achieve. On the other hand, at higher operating frequencies,
Fresnel zones are shorter. Also, as mentioned earlier, number of Fres-
nel zones that reach the receiving antenna unobstructed together with
the reflectivity of the path surface decide the signal strength a t the
receiving antenna. It may be recalled that nulls correspond to even
numbered zones while odd numbered zones correspond to peaks. It can
also be observed that with a large number of zones unobstructed and a
low reflective surface along the path length, the attenuation in the path
is more or less equal to the free space loss. The effect of number of
Fresnel zones travelling unopposed on the signal strength a t the
receiver can also be illustrated with the help of Fig. 12.14 which shows
the cross-section of the wavefront close to the receiver. In the figure
shown, even the fourth Fresnel zone is travelling unopposed. Now if
the heights of the transmitter and receiver towers are lowered, some
portions of the wavefront is likely to be blocked either by the earth's
curvature or an obstacle in the path. If the canceling zones are blocked,
the signal a t the receiver will increase. On the other hand, if in-phase
zones are obstructed, the signal strength will reduce. As an example,
when the obstruction reaches the edge of the first Fresnel zone, the
signal will be twice that in the free space. Raising or lowering the
heights of the towers a t this point will reduce the signal strength. The
depth of the nulls obtained will depend upon the reflectivity of the
obstacles. A highly reflective obstacle will give deep nulls. If a large
number of zones are cleared, an obstacle a t a n appropriate location can
also cause deep nulls.
MICROWAVE COMMUNICATION LINK BASIC DESIGN 603

Fig. 12.14.

Another interesting point here is that as a t higher operating fre-

quencies, Fresnel zones are closely spaced, a slight increase or decrease
in antenna heights can even remove the cancellation. Use of a nar-
rower beam antenna can be another possible remedy to the cancella-
tion problem. Yet another method of reducing the cancellation is to tilt
the antennas slightly upwards. This can however be used only if the
loss of power that it would cause is tolerable.
Multi-Path Fading and Outage Considerations
Having designed the path for a certain Fresnel clearance, the next
main issue of concern to the designers is the 'Path Availability' or the
'Outage Time'. Two main causes that affect the Outage Time are the
Rain Induced Fading and the Multi-Path Fading also called Dispersive
Fading. The reason why multi-path fading is also known as dispersive
fading is the following. One of the mechanisms that causes multi-path
fading is refraction in the atmosphere that results in signal arriving at the
receiving antenna via multiple paths. Due to different distances involved
therein, the signals arrive at the receiver with varying phase. If the two
signals travelling two different paths arrive out of phase, there is a deep
cancellation. Thus there is a frequency dependent attenuation charac-
teristic. It is for this reason of fkequency selective attenuation that it is
called dispersive fading. The attenuation characteristics in case of disper-
sive fading looks like the one shown in Fig. 12.15. The exact shape of the
notch depends upon the relative delay and attenuation of the two paths.
The depth of the notch is determined by the difference in the levels of the
two signals whereas the frequency at whch this notch occurs is decided by
the difference in phase. In addition to causing a notch in the attenuation
characteristics, multi-path fading also causes distortion in the Differential
Delay characteristics.
While rain induced fading is the predominant factor for analogue
communication above 10 GHz, digital transmission a t these frequen-
cies are more affected by multipath fading due to high data rates in-
volved therein. Due to significant effect of Dispersive Fading
particularly on the digital systems, a concept of Dispersive Fade Mar-
gin (DFM) has evolved. This allows the designer to calculate the outage
time due to selective fading simply by substituting DFM in piace of
Fade Margin in the expression discussed earlier to compute the outage
time. This has allowed the designers to use the existing path design
programmes to be easily modified for use on digital system paths.

-20 -15 -10 -5 0 5 10 15 20

f (MHz)-
Fig. 12.15.
Of course, the significance of DFM depends upon its relative value
as compared to normal system fade margin often called Flat Fade Mar-
gin (FFM). The two type of fade margins are combined into a single
expression to give what is known a s Composite Fade Margin (CFM)
that can be computed from the following expression.
CFM = - 1 0 log [ l o - FFM'10 + 10- DFM'lO]
Fig. 12.16 shows a plot of CFM versus FFM for three different
values of DFM equal to 2 0 , 3 0 and 4 0 dB. It can be seen from the plot
that an improvement in CFM occurs as long a s FFM is less than or
equal to the DFM. Once FFM exceeds, let us say, beyond (DFM + 6 ) dB,
there i s very little further improvement. The outage or the path
availability can now be computed from
U = a x b x 2 . 5 x 10-6xfx~3x10-CFM'10

10 15 20 25 30 35
FFM (dB)

Fig. 12.16
- 40 45 50 55 60
MICROWAVE COMMUNICATION LINK BASIC DESIGN 605

Problem 12.10 : A certain single hop microwave link has a Disper-

sive Fade Margin (DFM) specification of 40 dB. Determine its Com-
posite Fade Margin (CFM) i f i t has Flat Fade Margin (FFM) o f 3 0 dB.
Solution : DFM = 40 dB, FFM = 30 dB
Substituting these values in the expression for CFM,
CFM = - 10 log [lo- FFM/10 10- DFM/lOl
+

= - 10log [lo- 3 ~ 1 +0 10-4 ~ / 1 0 ~

= - 10log [lo- 3 + 10-4~
=- 10 log [(1/1000) + (1/10000)1
= - 10 log (11/10000) = - 10 log (0.0011)
= - 29.58 dB

Problem 12.11 :I n case of Problem 12.10, prove that there is prac-

tically no improvement i n CFM value when DFM i s increased from 40
db to 50 dB.
Solution :DFM = 50 dB
CFM can now be computed from :
CFM = - 10 log [lo- FFM/10 10- DFM/1O1
+

=-10log [lo- 3 + 10-51

= - 10 log [(1/1000) + (1/100000)1
= - 10 log [0.001+ 0.000011
= - 10 log (0.00101) = 29.95 dB
Thus, the CFM increases by only 0.37 dB for a 10 dB increase in
DFM which is very marginal.

Problem 12.12 :A microwave link has DFM and FFM specifica-

tions o f 4 0 a n d 3 0 dB respectively. Determine the 'Outage Time' in
minutes per year for a path length of 10 miles a n d a n operational fre-
quency of 2 3 GHz. Assume a = 1 and b = 0.5.
Solution : For given DFM and FFM values, CFM has already been
computed in Problem 12.10 to be equal to 29.58 dB. Path Unavailability
or the Outage in decimal form can be computed from :
U = a x b x 2.5 x 10- x f x D~ x 10- CFM/10
Substituting the values,

U (in minutes per year)

= 0.00003167 x 525600 = 16.64 minutes per year
606 MICROWAVES AND RADAR

12.3. Analog and Digital Microwave Links

P e r f o r m a n c e Comparison.
The salient features of the two t: pes of systems are briefly outlined
below :
Analog Microwave Systems ;The salient features include
1. Analog Microwave Systems offer superior voice channel per-
formance in small to medium sized links. In a n analog sys-
tem, voice channel performance remains usable (above 15 dB
Average Voke Power to Noise ratio) even when the received
signal level has faded below the level which would have
caused data channel outage.
2. Serious consideration needs to be given 'to the assignment of
baseband frequencies in case of a n analog system.
3. For most system sizes, analog system offers performance
advantage for voice channel data modem users.
4. For given bandwith and voice channel capacity, analog
system, when compared to a digital system, offers much
higher system gain which ie nothing but the Transmitter
Power to Receiver Sensitivky Ratio in dB. This is mainly
due to the poorer micro.?ave receiver threshold of the
bandwith efficient digital microwave system. Higher sys-
tem gain is generally considered a s a desirable feature
since i t is related to the zbility of the system to withstand
undesirable microwave signal propagation effects.
5. For given system g a i ~and bandwith, analog system' offers
higher voice channel capacity.
Digital Microwave Systems :The salient features include :
1. In a digital system, both voice and data channels are affected
similarly by microwave signal fading to t h e receiver
threshold. The performance below the threshold is unsatis-
factory for both voice and data.
2. There is no base band slot allocation difference to be considered
in case of a digital system. The digital system equivalent to base
band frequency slot allocation is time slot allocation, which
makes no-diffeience to the channel performance.
3. Digital systems offer definite performance advantage for voice
users in large systems and also for high-speed data.
4. For similar bandwidth and voice channel capacity, digital sys-
tem offers much higher data capacity. For similar data capacity,
it is much more bandwidth efficient for data transmission.
5. For given System Gain and Bandwidth, a digital system
offers higher data capacity.
MICROWAVE COMMUNICATION LINK BASIC DESIGN 607

12.4. Microwave System Specifications

In this section, we shall briefly describe both the microwave sys-
tem specifications and also the microwave equipment specifications.
The characteristics or the specifications that will be in focus include the
following :
1.Area of coverage
2. Number of hops and length of each hop
3. System Gain
4. Transmitter Power Output
5. Receiver Sensitivity
6. Noise Power Ratio
Area of Coverage
Area of coverage is one of the most important system charac-
teristics that affects both the system performance as well as system
reliability. It also affects the system availability due to outages due to
equipment failure since the area of coverage determines the amount of
microwave equipment required and no equipment is reliable hundred
percent..
Number of Hops and Length of Each Hop
The length and number of microwave hops in the system has a
direct bearing on system specifications such a s Signal-to-Noise
Ratio, Distortion (in a Voice Channel), Bit Error Rate (in a Data
Channel) and Availability.
System Gain
System Gain is the decibel difference between Transmitter Power
and Receiver Sensitivity. In fact, the microwave equipment perfor-
mance parameters of transmitter power output and receiver sensitivity
have been combined into a single specification in System Gain. Higher
transmitter output power or better receiver sensitivity or both give a
higher system gain. It is desirable to have a higher system gain as it
may allow longer hops, smaller antennas, improved signal-to-noise
ratio and less frequent outage due to multi-path fading.
Increased system gain allows longer hops. The free space
path loss and the path length have an inverse square law relationship
i.e. Path Loss is inversely proportional to square of the path length. In
order to have a certain minimum received signal level, increased sys-
tem gain may be used to allow longer hops. For a given total link path
length, longer hops implies reduced number of hops which means
reduced microwave equipment. The inherent and unavoidable failure
rates of cascaded equipment are accumulative. Therefore, reduced
equipment enhances reliability. It may also be mentioned that Mean
608 MICROWAVES AND RADAR

Time Between Failure (MTBF) of the basic multi-hop microwave sys-

tem is inversely proportional to the number of hops.
Increased System Gain also leads to use of smaller antennas in case
longer hops are not desired. Antenna Gain can be computed from :
Antenna Gain (in dB) = 20 Log f + 20 Log D - 52.6
Here, 'f is in MHz and D (Antenna Diameter) is in feet.
Antenna Gain is directly proportional to square of the antenna
diameter.
Higher system gain allows use of minimum size dish antenna. A
smaller a n t e n n a h a s associated advantages of being less costly,
reduced stress due to wind loading on the supporting structure. This
reduces overall system cost and increases systeri reliability.
- Increased system gain also leads to improved signal-to-
noise ratio. Increased system gain increases the signal level arriving
a t the microwave receiver. This improves S/N ratio in a voice channel
or the Bit Error Rate (BER) in a Data Channel if the intrinsic noise is
not the limiting factor.
Transmitter Power Output
Increase in transmitter power output increases system gain (from
the definition of system gain) which is a desirable feature a s seen in the
earlier paragraphs. Yet, i t is usually preferred to achieve increased
system gain by implementing improvements in receiver sensitivity.
The reasons for this are the following. Transmitter power enhance-
ment is costlier to implement. Also, cost of transmitter output power is
strongly dependent upon frequency of operation and it costs much more
a t higher operating frequencies. However, having reached a limit in
receiver sensitivity improvement, the transmitter power may be in-
creased, if needed. And in future, it is likely to happen with GaAsFET
technology.
Higher transmitter power often leads to a less reliable system. On
the other hand, low power is not only less costlier, the associated
reduced parts count, decreased equipment temperature etc. improve
overall system reliability. As a n illustration, if the MTBF of the trans-
rpitter with the power amplifier (usually the case with a higher output
power set-up) is known, the MTBF of the transmitter without power
amplifier can be determined from :
MICROWAVE COMMUNICATION LINK BASIC DESIGN 609

where [ ]
= MTBF of the transmitter without power
(Ml"BF)l amplifier
= MTBF of the power amplifier
[(MTBFh]

= MTBFaf the transmitter with power amplifier

[(MTBFJ
Receiver Sensitivity
Receiver Sensitivity is the main specification of the receiver and
it indicates the weak signal performance of the receiver. As already
outlined in the earlier paragraphs, improvement in receiver sensitivity
is always less expensive to achieve when compared to achieving enhan-
cement in transmitter power. Therefore, it is a preferred choice when
it comes to improving System Gain specif cation. Of course, improve-
ment in receiver sensitivity is also limited by Random Noise (the KTB
noise). The limiting value of the receiver noise figure is 0 dB. Contem-
porary microwave receivers are a dB or two away from this ideal value.
It is not recommended to design systems with significantly higher
received signal levels to improve (S/N) ratio than is specified for the
microwave equipment. This may degrade Receiver NPR (Noise Power
Ratio), discussed in the following paragraphs, due to receiver overload
or distortion.
Noise Power Ratio
Noise Power Ratio (NPR) is a parameter that indicates the over-
all performance of the multiple channel microwave system. I t is
measured in decibels and a larger number is better. I t is a major con-
trolling factor in system design and system cost. It is a partly a function
of system gain and partly dependent upon received microwave signal
level. Higher NPR primarily allows improved (S/N)and lower distor-
tion in a voice multiplex channel on a microwave system. Higher NPR
also gives a lower Bit Error Rate on data channels. The basis of NPR
measurement is that microwave systems carry multiple communica-
tions channels simultaneously. These multiple channels are affected
not only by noise floor of the transmitter and receiver but also by a
noise like interference above this level dce to harmonic and inter-
modulation distortion products. NPR is measured as follows. In the
NPR measurement test set up, the multiple channel base band signal
is represented by a noise generator set for equivalent power level and
frequency spectrum. The noise power is measured a t the system output
for channel under test. The noise generator spectrum is then notched
a t the frequency of channel under test. Residual noise power is
measured a t system output in channel under test. The ratio of the two
610 MICROWAVES AND RADAR

readings is the NPR. NPR measurement includes all sources of channel

degradation such a s Transmitter Idle Noise, Receiver Idle Noise,
Transmitter Thermal Noise, Receiver Thermal Noise, Transmitter Dis-
tortion and Receiver Distortion. ( S N ratio limited by Idle and Ther-
mal noise could be improved by increasing base band signal level. But
increasing the base band level also increases the distortion level which
tends to degrade NPR. There is a n optimum base band level which is
sufficiently high to gain maximum improvement over intrinsic and
thermal noise without creating' a n excessive distortion. Fig. 12.17
shows the plot of NPR versus channel signal level. NPR increases with
signal level up to a certain point called the Optimum Point beyond
which it falls.
Channel Signal level (dB).--,
r

Fig.12.17.
It may be noted that if idle and inter-modulation distortion could
be eliminated completely, the NPR could improve by several decibels
given that base band levels are fixed by bandwidth considerations and
received signal levels are a t the levels essentially governed by receiver
overload-considerations.

Problem 12.13 : Acertain microwave transmitter that uses a chain

ofpower amplifiers at its output has an MTBFfigure of 20000 hours. If
the power amplifier portion alone has an MTBF of 60000 hours, deter-
mine the improvement in the MTBF ofthe transmitter if it did not use
a power amplifier.
Solution :The MTBF of the transmitter without power amplifier
can be computed from :

[A]I-&[ I&[
[A] =
MTBF = 30000 hours
= =
1

MTBF improves by 10000 hours.

MICROWAVE COMMUNICATION LINK BASIC DESIGN 611

Review Questions
1. Briefly describe the effect of Fresnel Zones on the strength of
the received signal. How does reflectivity of the path surface
aid or oppose deterioration of signal strength ?
2. Compare the performance of a single hop and multi-hop
microwave links on the basis of Path unavailability or Outage
time.
3. How does SPACE DIVERSITY help in improving the Path
availability or reducing the Outage time ? What is the typical
optimum spacing between the two receiving antennas ?
4. How do t h e under-mentioned p a r a m e t e r s affect the
microwave link performance ?
( a ) Transmitter Power Output
( b ) System Gain
( c ) Receiver Sensitivity
(dl Noise Power Ratio
5. ( a )Why is it preferred to improve the system gain specifica-
tion by improving the receiver sensitivity rather than in-
creasing the transmitter power output ?
(c) How does slight tilting of transmitting and receiving an-
tennas in the upward direction help in reducing amount of
reflection ?
6. Briefly describe the following in the context of a microwave
communication link :
( a ) Fade Margin ( b ) Outage Time
( c ) Multi-Path Fading ( d ) Composite Fade Margin
7. Briefly compare the performance of Analog and Digital
microwave systems on the basis of System Reliability, Outage
Time and Noise Performance.
8. With the help of relevant expression, explain how Outage
Time depends upon Path length, Operating frequency, Fade
margin and Climatic and Topographic conditions.

Problems
1. A single hop microwave system with path length of 25 miles
operates a t 10 GHz. Determine first Fresnel Zone distance a t
a location 10 miles from the transmitter. (43.3 ft)
2. With the help of relevant expression, prove that Composite
Fade Margin (CFM) remains unchanged even if DFM and
FFM figures are interchanged.
3. If the first Fresnel Zone clearance for a certain microwave
link is 50 feet, determine clearance required for fourth Fres-
nel Zone. (100 ft)
612 MICROWAVES AND RADAR

4. For the microwave system of Problem-1, if the transmitter

tower height is 143 feet, d e t e n n i ~ etha distance from trans-
mitter u p to which there will be first Fresnel Zone clearance.
Assume straight line propagation. (10 miles)
5. Assuming an earth bulge condition a r d a K-factor of 2/3, what
would be the answer in case of prcblem-4 ? (5.8 miles)
6. If the frequency of operation of a single hop microwave link
is doubled and path length halved, what would be its effect
on unavailability factor if other parameters including Fade
margin remain unchanged?
(Unavailability Factor reduces by a factor of 4 )

SELF EVALUATION EXERCISE

Multichoice Questions
1. Effect of earth's curvature is maximum a t
(a)25% of path length from transmitter
( b ) mid way between path length
(c) 75%of path length from transmitter
( d ) the receiver
2. For a given path surface reflectivity, odd number Fresnel
Zone clearance
(a) increases signal strength a t the receiver
(b) reduces signal strength a t the receiver
(c) produces a very deep null
(dl produces a signal peak
3. For a given path surface reflectivity, even number Fresnel
Zone clearance
(a) increases signal strength a t the receiver
(b) produces a signal peak a t the receiver
(c) produces a deep null a t the receiver
(dl decreases signal strength at the receiver
4. While using space diversity,
(a)two transmitting antennas spaced 30 to 60 feet in the
vertical direction are used.
(b) two receiving antennas spaced typically 30 to 60 feet in
the vertical direction are used.
(c) multipath signals are allowed to combine in phase a t
the receiver.
(d) none of these
5. A DFM (Dispersive Fade Margin) of 20 dB and FFM (Flat
Fade Margin) of 30 dB yield a CFM (Composite Fade Margin)
MICROWAVE COMMUNICATION LINK BASIC DESIGN 613

figure of 19.6 dB If the values of DFM and FFM are inter-

changed, new CFM would be
(a) 19.6 dB only (b) 39.2 dB
(c) 9.8 dB ( d )none of these
6. A single hop microwave link has an outage time of 10 minutes
per year. A multi-hop link comprising of four such identical
hops will have a worst case outage of
(a) 160 minutes per year (b) 150 seconds per year
(c) 40 minutes per year (d) 10 minutes per year only
7. If FFM (Flat Fade Margin) is greater than DFM (Dispersive
Fade Margin) by say more than 10 dB, then CFM (Composite
Fade Margin)
(a) Equals FFM (b) Equals DFM
(c) IS half of DFM (d)Is half of FFM
8. Ducting phenomenon
(a) occurs when the refractive index starts increasing with
height
( b ) helps microwaves propagate beyond line-of-sight
(c) occurs when refractive index varies very rapidly over a
certain region in atmosphere caused by certain condi-
tions in the atmosphere leading to creation of a tempera-
ture inversion layer
(d) both (b) and (c) are correct
9. Cne of the following is not a n advantage of a n analog
microwave system
(a)Analog microwave system offers superior voice channel
performance
(b) For a given system gain and bandwidth, analog system
offers higher voice channel capacity
( c ) For given bandwidth and voice channel capacity, analog
system offers higher system gain
(dl There is no base band slot allocation difference to be
considered in case of an analog system
10. System gain is defined as
(a) ratio of transmitter output power to receiver sensitivity
( b ) ratio of receiver sensitivity to transmitter power output
(c) product of transmitter output power and transmitting
antenna gain
(d) both (a)and (b) are correct
11. One of the following is an advantage of a higher system gain
(a) It allows longer hops
614 MICROWAVES AND RADAR

( b ) It allows smaller antennas

(c) It leads to improved S/N ratio
(dl It reduces system cost
12. Mark the incorrect expression :
( a ) CFM = - 10log [ I O - ~ +~10-DFM/lO]
~ / ~ ~
( b ) CFM = - 20 log [lO-FFM/10+ 10-DFM/lO]
(c) U =~ X D ~10-F/10X
( d ) FN= F1

ANSWERS TO MULTI-CHOICE QUESTIONS

 ANSWER TO
SELF EVALUATION EXERCISE
-
Chapter 1 :Introduction to Microwaves
Multichoice Questions
1. ( a ) 2. ( b ) 3. (c.) 4. ( a )
5.(b) 6.b) 7.(d) 8.(a)
9. ( a ) 10. ( a )
Fill in the blanks :
( i )30 H z to 300 Hz ( i i ) 100 k m to 10 krn
(iii)hlF (iv)UHF ( v )Near infra-red

-
Chapter 2 :Maxwell's Equations
Multichoice Questions :
1. ( a ) 2. (d) 3. (a) 4. ( b )
5. ( c ) 6. ( c ) 7.(a) 8. ( c )
9. ( b ) 10. ( c ) 11. ( a ) 12. (c)
13. ( a ) 14. ( a ) 15. ( b ) 16. ( a )
17. ( b ) 18. ( d ) 19. ( b ) 20. ( a )

Fill in the blanks :

1. V .A 2. scalar 3. scalar 4. can
5.can
6. ( i )normal (ii)continuous 7. (i)normal
(ii)continuous 8. (i)tagential (ii)current
9. non-existence 10. maximum 11. maximum
12. can not 13. V X A 14. curl of H
15. capacitor

Match the Following :

(1-3-3) (24-1) (3-1-4) (4-2-2)

Chapter-3:Transmission Media
Multichoice Questions :
1. ( c ) 2. ( a ) 3. ( c ) 4. ( b )
5. ( a ) 6. ( b ) 7 . (e) 8. ( a )
616 MICROWAVES AND W A R
9. (b) 10.(a) 11. (a) 12. (c)
13. (c) 14.( a ) 15.(b) 16. (a)
17. (b) 18. (c) 19. ( b ) 20. (c)
21. (a) 22. (c) 23. ( c ) 24. ( b )
25. ( c ) 26.(b) 27.(a) 28.(b)
29. 30. (b) 31. ( a ) 32.(c)
33. !b) 34. ( a ) 35. (c) 36. (b)
t". (31 38. ( c ) 39. ( d ) 40. ( b )
bk. (3) 42.(a! 43.(c)
46. ( b )
AZatch the Following :
11-31 (2-4) ( 3 - 1 )

Chapter4 :Microwave Components

Multichoice Questiom :
1. (a) 2. ( a ) 3. ( d ) 4. (d) 5.(a)
6.k) 7.(a) 8.(a) 9.(E) 10.(c)
11. ( a ) 12. (b) 13. (c) 14. ( b ) 15.( a )
16.( c ) 17.(dl 18. u 19. (d) 20.(b)
21.(L) 22.( d ) 23.(a) 24. (d) 25.( a )

-
Chapter 5 :Microwave Tubes
Multichoice Questions :
1.( a ) 2.(c) 3. (a) 4. (c) 5. (c)
6. (c) 7.(dl 8.a 9. (a) 10.(b)
11. ( d ) 12. (6) 13. (b) 14.( a ) 15. (b)
16. (c) 17.( a ) 18. ( h ) 19. (c) 20. (c)
(1-b) !2u) (3d) (4*) (5e)
(6c) (74

-
Chapter 6 :Semiconductor Microwave Devices
Multichoice Questions :
1.(dl 2.(a) 3.( a ) 4.( a ) 5. (b)
6. ( a ) 7.(a) 8. (b) 9. (d) 10. ( d )
ANSWERS TO SELF EVALUATION EXERCISES 617

1. Gunn diode 2. IMPATT diode

3. GaAs MESFET 4. Parametric Amplifier
5 . Step Recovery Diode 6. MASER
Fig. 6.69 ( a ) Tunnel diode
Fig. 6.69 ( h ) Gunn diode
Fig. 6.69 ( c ) BARITT diode
Fig. 6.69 ( d ) PIN diode
Fig. 6.69 ( e ) Backward diode

-
Chapter 7 :Antennas
Multichoice Questions :
1.(d) 2. ( b ) 3.( a ) 4. (b) 5. ( a )
6. ( c ) 7. ( a ) 8. ( a ) 9. (c) 10. ( c )
11. ( d ) 12. ( a ) 13. VI) 14. ( a ) 15. ( a )

-
Chapter 8 :Microwave Measurements and
Instrumentation
1.( b ) 2. ( a ) 3.( b ) 4 . (d) 5. ( a )
6. ( c ) 7. ( a ) 8. ( a ) 9. (d) 10. ( b )

Chapter - 9 :Radar Fundamentals

Multichoice Questions :
1. ( b ) 2. ( a ) 3. (a) 4 . (a) 5. ( b )
6 . (d) 7. ( a ) 8. ( c ) 9. ( b ) 10. ( b )
11.( b ) 12.( a ) 13.(c) 14. (d) 15. (c)
16. ( b ) 17. ( a ) 18. (c) 19. ( c ) 20. ( c )
True 1 False Statements
1 . True 2.False 3. False 4. False 5. True
6.True 7.True 8.True 9.True 10.False

-
Chapter 10 :Satellites and Satellite Communication
Multichoice Questions :
1. ( d ) 2.( a ) 3. ( a ) 4 . (a) 5. (a)
6..(c) 7. ( e ) 8. ( e ) 9. (g) 10. ( a )
618 MICROWAVES AND RADAR

11. ( b ) 12. ( b ) 13. ( a ) 14. ( b ) 15. ( a )

16. ( a ) 17. ( c ) 18. ( c ) 19. ( a ) 20. ( b )
21. ( a ) 22. ( b ) 23. ( d ) 24. ( a ) 25. ( a )
26. ( a ) 27. ( a ) 28. ( c ) 29. ( b ) 30. ( b )
Incomplete Statements :
1. Arthur C. Clarke 2. Kepler's Laws
3. Sun synchronous 4. Decreases
5. Narrower 6. Remain unaffected
7. At one of foci of ellipse 8. FDMA

True 1 False Statements :

1. True 2. True 3. False 4. False 4. False
5. True 6. False 7. True 8. True.

-
Chapter 11 :Electronic Warfare
Multichoice Questions :
1.(dl 2 . (c) 3. ( b ) 4. ( d ) 5. (b)
6. ( b ) 7. ( a ) 8. ( a ) 9. ( c ) 10. ( b )
 INDEX
Code division multiple access (CDMA).
5 14
Coherent transmitters, 424
Analog microwave systems, 606 Command guidance, 481
Angular position accuracy, 405 Conical scan, 462
Antenna aperture, 302 Continuous count counters, 378
Antenna gain 294,-295 Counter specifications, 378
Antenna impedance, 301 Coupling loops, 115
Array feed reflector antenna, 319 Coupling probes, 113
A-scope, 438 Cross field amplifier, 197-199
Attenuators, 136-139 Cross-range resolution, 407
Azimuth and elevation angles, 502-503 CW radar, 442-443
Cylindrical magnetron, 190

Backward wave oscillator, 204 Deception jamming, 574-576

Bandwidth-antenna, 298 Decoys, 579-580
Barrage jamming, 573 Degenerate parametric amplifiers, 225
BARRITT diode, 252-254 Demand division multiple access, 514
Beam rider guidance, 479 Dielectric phase shifter, 140
Beam width-antenna 297 Digital microwave system, 606
Biot Savart law, 22 Dipole antenna, 308
Bistatic radar, 476 Direct digital synthesis, 385-387
Blind speeds in MTI, 452 Direct frequency synthesis, 382
Broadband Parametric amplifiers, 228 ~ i ~ ~ coupler,
~ t125-130
i ~ ~ ~ l
B-scope, 439 Directional pattern, 296
Directivity, 295
Displacement currect, 22
Doppler ambiguity, 403
Doppler granularity, 403
Cable television, 517 Doppler resolution, 408
Capacitive windows, 117 Duplexer, 395
Cassegrain antenna, 318
Cavity modes, 142
Cavity resonators, 141-146
Cavity wavemeter, 148
Chaff, 578-579
9
Earth bulge, 589
Choke flange, 135 Earth coverage angle, 504
Circulator, 133 Electrical length of antenna, 307
Circular waveguides, 94 Electromagnetic wave spectrum, 2
- cut-off wavelength in, 94 Electronic counter countermeasures
- power loss in, 95 (ECCM), 581-584
- Propagation of TE waves in, 89 Electronic countermeasures (ECM),
- propagation of TM waves in, 93 556,571-580
 MICROWAVES AND RADAR

Electronic support measures (ESM),

555-568
Electronic warfare, 554-584
Electronic warfare-current trends, 584 Idler frequency, 225
Electro-optic counter measures Impedance matching, 55
(EOCM), 580-581 IMPATT ciiode, 248-252
E-plane tee, 119 Indirect synthesis, 383
Inductive windows, 116
Insertion loss, 344
Isolators, 130

False alarm number, 399

False alarm rate, 400
Faraday's law, 20
FFT spectrum analyser, 373 Jamming techniques, 571-577
FM-CW radar, 443-449
Focal point feed reflector antenna, 315
Fresnel zone clearance, 602
Frequency counters, 375-379
Frequency division multiple access Klystron, 167-174
(FDMA), 513
Frequency pushing and pulling, 197
Frequency synthesisers, 381-387
Frequency wavelength relationship, 4
F-scope, 440 Laser, 261-265
Laser radar, 479

1 Laser warning receivers, 563-566

Lens antenna, 321
Lobe switching, 461
Geostationary orbit, 495 Log periodic antenna, 325-326
Geosynchronous orbit, 495 Loop direction coupler, 128
Gridded tubes-limitations of, 166 LSA mode in Gunn diode, 247
GUNN diode, 242-248
Gyrotron, 206

Magic tee, 121

Magic tee-Applications of, 123
Helical antenna, 324 ~ a g n e t r o n185-197
,
HEMT, 257 Marconi antenna, 308
Hertz antenna, 307 Maser, 259-261
HFET, 257 Maxwell's equations, 11-29
Homing guidance, 481-482 MESFET, 256
Horn antenna, 322 Meteorological satellites, 522
H-plane tee, 120 Microstrips, 152
Hybrid junctions, 121 Microwave components, 112-156
Hybrid MICs, 373 Microwave FETs, 255-258
Hybrid ring, 149 Microwave frequency bands, 3
Microwave frequency counters, 379
Microwave Integrated Circuits (MICs),
273-280
Microwave link design, 588
Microwave path design, 602 Parallel plate magnetron, 187
Microwave power measurement, Parametric amplifiers, 225-228
354-358 Path clearance, 588
Microwave signal propagation, 4 P a t h profiles, 593
Microwafe system specifications, 607 Phase noise, 363-365
- area of coverage, 607 Phase shifters, 139-141
- - noise power ratio, 609 Phased array antenna, 325-331
- number of hops and length, 607 Phased array radar, 325
- receiver sensitivity, 609 Physical length of antenna, 307
- system gain, 607 PIN diodes, 228-237
- transmitter power output, 608 Point contact diode, 237
Microwave tubes, 165-208 Polarisation-Antenna, 298
Microwave-Advantages of, 6 PPI display, 440
Microwave-Applications of, 7 Primary radar, 477
Missile approach warning receivers, Probability of detection, 399
566-568 Probability of false alarm, 399
MITATT diode, 255 Probability of noise, 399
Molniya orbit, 496 Pulse compression radar, 468-471
Monolithic Microwave Integrated Pulse doppler radar, 456
Circuit, 280-283 Pump frequency, 225
Monopulse tracking, 463-466
Monostatic radar, 476
Moving Target Indicator (MTI), 449-453
Multicavity klystron, 174-177
Multipath fading, 594, 603 Quasi-coherent transmitters, 425
Multiple access systems, 513-515

Radar classification, 441-442

Network analyser, 344-354 Radar displays, 438-440
Noise jamming, 571-574 Radar functions, 398
Noise measurement, 358-362 Radar guidance of missiles, 479-482
Noise power ratio, 609 Radar range equation, 412-416
Non-coherent transmitters, 427 Radar receivers, 429-431
Non-degenerate Parametric amplifiers, Radar transmitter parameters, 427
227 Radar transmitter, 424-429
Non-resonant antennas, 306 Radar warning receivers, 559-563
Radar waveforms, 416-423
Radiation resistance-Antenna, 302
Range accuracy, 404
Range ambiguity. 401
Offset feed reflector antenna. 317 Range quantisation and granularity, 402
Orbit period, 501 Range resolution, 406
Orbit velocity. 501 Range tracki'lg, 466-467
Over-the-horizon-radar (OTHR), Receiver sensitivity, 609
474-475 Receiver parameters, 431-434
 MICROWAVES AND RADAR

Reciprocal counters, 377 Stealth technology, 568-571

Rectangular waveguides, 77 Step recovery diode, 224
- power loss in, 87 Strip lines, 151
- transverse electric modes in, 77 Sun synchronous orbit, 496
- transverse magnetic (TM) waves Synthesised function generators, 386
in, 85 Synthetic aperture radar, 471-473
- various TE modes in, 84
- various TM modes in, 86
Reflection coefficient, 343
Reflection parameters, 343-344
Reflectometer, 129
Reflector antennas, 312-319 TE modes in rectangular waveguides, 84
Reflex klystron, 177-185 Thick film hybrids, 276
Remote sensing satellites, 526 Thin film hybrids, 279
Resonant antennas, 306 Time division multiple access (TDMA),
Resonant modes in magnetrons, 196 5 14
Resonant windows, 117 Time domain reflectometry, 366-367
Return loss, 344 Tracking radar, 459-466
Rhombic antenna, 310 Transferred electron effect, 242-244
Rotary phase shifter, 139 Transit time devices, 248
Rotating joints, 136 Transit time effects, 134-135
Transmission coefficient, 344
Transmitter power output, 608
Transmission lines, 48-98
Transmission parameters, 49, 344
Sampled sine synthesis, 385 - equivalent circuits, 48
Satellite data communication services, - sources of losses in, 48
5 19 - types of, 53
Satellite eclipses, 506-507 Transponder, 511
Satellite frequency bands, 491 TRAPAIT diode, 252
Satellite orbits and trajectories, 491-495 Traveling wave tube, 201-204
Satellite telephone services, 519 True and Apparent ranges in radar, 457
Satellite television, 515 Tuning posts and screws, 118
Satellites for military applications, 533 Tunnel diode, 239-241
Satellites for scientific applications, 529 TUNNETT diode, 254-255
SAW devices, 153
Scalar network analyser, 345
Scattering parameters, 341-343
Schottky barrier diode, 238-239
Secondary radar, 478 Universal counters. 375
Sequential lobbing, 462
Single point to multipoint connectivity,
515
Slotted line, 150
Slow wave structure, 203
V-antenna, 310
Smith chart, 58
Varator, 222-224
Space diversity, 597
Vector algebra, 12-19
Spectrum analyser, 367-373
Vector network analyser, 345
Spot jamming, 573
Velocity modulation, 168
Standing wave ratio, 344
Velocity tracking, 468
INDEX 623

Wye circulator, 105

Waveguide flanges, 134-136
Waveguide junctions, 119-124
Waveguides, 62
- modes, 63
- parameters, 65
- plane wave propagation, 71 Yagi-Uda antenna, 309
- transverse electric modes in. 77