Novel Approaches to the Design of Phased Array Antennas

By

Danial Ehyaie

A dissertation submitted in partial fulfillment

of the requirement for the degree of
Doctor of Philosophy
(Electrical Engineering)
in The University of Michigan
2011

Doctoral Committee

Professor Amir Mortazawi, Chair

Professor Mahta Moghaddam
Professor Kamal Sarabandi
Associate Professor Jerome P. Lynch
© Danial Ehyaie 2011
All Rights Reserved
 

To the most beloved people in my life,

My parents: Reza and Minoo

My sister and brother: Nona and Dana

My grandmother and uncle: Zarrintaj and Mehran

ii 
 
 

TABLE OF CONTENTS

DEDICATION……………………………………………………..…………....... ii

LIST OF FIGURES……………………………………………………................. v

LIST OF TABLES……………………………………………………................... xi

ABSTRACT…………………………………………………….……………..….. xii

CHAPTER

1. Introduction ………………….…………………………………..……............. 1
1.1. Background………………..……………….……..…...…...….…..….….… 2
1.2. Emerging Applications of Phased Array...……….………..................….… 4
1.3. Phased Array Principle……………………………………..…….….…….. 10
1.4. Advantages of Using Phased Arrays……………………………….…........ 13
1.4.1. Spatial Filtering………………………...………………....……........ 13
1.4.2. Gain Boosting in Transmit Phased Array……………...………........ 16
1.4.3. Sensitivity Improvement in receive Phased Array………...…...….... 18
1.5. Phased array Architecture………………………….…...……...………...… 20
1.5.1. Phased Arrays Based on Feed Network………………..….…..….… 20
1.5.1.1. Parallel-Fed Arrays………………….....…..…................... 21
1.5.1.2. Series-Fed Arrays……………………….……..…….…..... 22
1.5.2. Phased Arrays Based on Phased Shifting Stage……...…….…..…... 23
1.5.2.1. RF Phase Shifting………………….…….........…….…..... 23
1.5.2.2. LO Phase Shifting…………………….…….....…………. 25
1.5.2.3. IF Phase Shifting…………………….…….…....……....... 27
1.6. Digital Phased array………………………….…………………...…...…… 28
1.7. Challenges in Design of Phased Arrays...……………..… ………….......... 29
1.8. Review of State of Art Phased Array Design……………...…………..….. 31
1.9. Overview of Thesis………………………………………...…...…..……… 35

2. A New Phased Array Design Based on Extended Resonance Technique...... 38

2.1. Introduction ………….………….…….………...….................................... 39
2.2. Extended Resonance Phased Array Concept……………..………….......... 40
2.3. A New Extended Resonance Circuit to Enhance Maximum Achievable
Inter-Element phase shift……………………………..……………...…….. 42
2.4. Modular Extended Resonance Phased Array….……………….……........... 52

iii 
 
 

2.5. Heterodyne Extended Resonance Phased Array………...…......................... 55

2.5.1. Overview of Heterodyne Modular Extended Resonance Phased
Array………………………………………………………………... 55
2.5.2. IF Circuit Design………………………..………………...……….... 58
2.5.3. LO Circuit Design……………………….…...................................... 60
2.5.4. Antenna Design…………..……………….……................................ 62
2.6. Fabrication and Measurement results…...……..…..…................................. 65
2.7. Chapter Conclusion….……..……………..................................................... 73

3. A New Bidirectional Series-Fed Phased Array………………….………...… 74

3.1. Introduction ……....………..……………...….............................................. 74
3.2. Overview of Phase Shift Requirement in Different Types of Phased
Array.............................................................................................................. 75
3.3. A New Bidirectional Series-Fed Phased Array to Reduce the Required
amount of Phase shift……..……………………..………………….……… 78
3.3.1. Bi-Directional Phased array Overview ….……..…………….…...... 78
3.3.2. Operation theory of Bidirectional Series-Fed Phased Array............... 80
3.3.3. Bidirectional Feed Network Design …….…...……….……….......... 80
3.4. A New Phase Shifter Design …………….…...…......................................... 87
3.4.1. Overview of Common Phase Shifter Design Approaches……..…… 88
3.4.2. Phase Shifter Design ………………...……………....…....... 91
3.5. Fabrication and Measurement Results ……....………..……..….……......... 94
3.6. Chapter Conclusion ……………………………....………..……....…......... 103

4. A Novel Design of Low Complexity Phased Arrays......…………………....... 105

4.1. Introduction ……..…..……...……………………………………………… 105
4.2. General Concept of Phased Array Using a Single Phase Shifter ………...... 107
4.3. Phased Array Architecture ……………………....………………………… 113
4.3.1. Feed network Design …….…………………..…………………...... 113
4.3.2. Phase Shifter Design …………………………..….……………....... 115
4.3.3. Symmetrical Phased array Design……………………..……............ 118
4.4. Fabrication and Measurement Results ……...……….…….………….…… 118
4.5. Chapter Conclusion ………………………………..…..…...………..…..... 121

5. Conclusion and Future Work ……………………........................................... 122

5.1. Thesis Summary ………………..…………..…………………...……….… 123
5.2. Future Work………………...…………..……………………………… 126
5.2.1. Extended resonance Phased Array………………………………….. 127
5.2.2. Bi-directional Series-Fed Phased Array……………………………. 128
5.2.3. Phased Array using a single phase shifter………………………….. 120

BIBILOGRAPHY…………………………………………………………..…....... 130

iv 
 
 

List of Figures  

Fig. 1.1. Multiple unit steerable antenna array developed by Friis and Feldman for short
range reception……………………………………………………………….. 3
Fig. 1.2. The Sea based X-band radar is a combination of the world’s largest phased array
radar carried aboard mobile, ocean going semi-submersible platform…….. 5
Fig. 1.3. Using radar sensors in the car can provide the driver with several driving-aid
functions……………………………………………………………………… 6

Fig. 1.4. SiBEAM 60 GHz phased array transceiver for wireless video-area network. 8

Fig. 1.5. HAARP is a research station directed by the Air Force Research Laboratory's
Space Vehicles Directorate in Gokona, Alaska to gather data about the
atmosphere……………………………………………………………………. 9

Fig. 1.6. Block diagram of N-element phased array…………………………………… 11

Fig. 1.7. Array factor for a linear 4-element phased array……………………………. 14

Fig. 1.8. Array factor for 4-element array with different antenna spacings….……….. 15

Fig. 1.9. Phased array transmitter can boost the gain in a desired direction….……….. 17
Fig. 1.10. Phased array receiver can improve the noise performance of the system… 18

Fig. 1.11. General diagram of parallel fed phased array…………………………….… 21

Fig. 1.12. General diagram of series fed phased array………………………………... 22

Fig. 1.13. General diagram of RF phase shifting phased array…………….…..………. 24

Fig. 1.14. General diagram of LO phase shifting phased array……….………….…… 26

Fig. 1.15. General diagram of IF phase shifting phased array………………………… 27
Fig. 1.16. Photograph and boding diagram for a 5-bit phase shifter………………….. 29
Fig. 1.17. Grouping antennas into a sub array each using one phase shifter can reduce the

v 
 
 

number of required phase shifters……………………………………….. 32

Fig. 1.18. Beam steering can be achieved by tuning the load reactances at
parasitic elements surrounding a central active element………………… 33

Fig. 1.19. Frequency scanning can be utilized to achieve beam steering …………….. 34

Fig. 1.20. Phased array using coupled oscillator array…….…………………………… 35

Fig. 2.1. The conventional design of phased arrays requires one phase shifter for each
antenna element………………………………………………………………. 39

Fig. 2.2. In an extended resonance phased array, phase shifters and power dividing
network are combined in a single entity……………………………………… 39

Fig. 2.3. The general structure of an extended resonance power dividing network…. 41

Fig. 2.4. The extended resonance circuit using just varactors as shunt susceptance… 43

Fig. 2.5. The proposed extended resonance circuit, with a block shown above, can
considerably increase the maximum achievable phase shift……………….. 44

Fig. 2.6. Maximum achievable inter-element phase shift versus the shunt inductance for
various varactors at 2 GHz…………………………………………………… 45

Fig. 2.7. Maximum achievable inter-element phase shift versus the shunt inductance for
various varactors at 2GHz……………………………………………………. 46

Fig. 2.8. The impedance transformers can be employed to facilitate the implementation
of the extended resonance circuit……………………………………………. 47

Fig. 2.9. Inter-element phase variation versus frequency for the 15 degree-scan range
phased array using different varactors………………………………………. 49

Fig. 2.10. Scan angle variation versus frequency for the 15 degree-scan range phased
array using different varactors……………………………………………….. 49

Fig. 2.11. Inter-element phase variation versus frequency for the 30 degree-scan range
phased array using different varactors………………………………………. 50

Fig. 2.12. Scan angle phase variation versus frequency for the 30 degree-scan range
phased array using different varactors………………………………………. 50

Fig. 2.13. Inter-element phase variation versus frequency for the 90 degree-scan range
phased array using different varactors………………………………………. 51

Fig. 2.14. Scan angle phase variation versus frequency for the 90 degree-scan range
phased array using different varactors………………………………………. 51

vi 
 
 

Fig. 2.15. The modular design of transmit extended resonance phased array………. 53

Fig. 2.16. The proposed phased array is a heterodyne-mixing structure that exploits the
extended resonance technique to achieve a uniform power distribution and to
control the phase progression across the phased array……………………… 57

Fig. 2.17. The IF extended resonance module at 2 GHz……………………………… 59

Fig. 2.18. The inter-element phase shift can be controlled by tuning the varactors
capacitances…………………………………………………………………… 60

Fig. 2.19. The LO extended resonance module at 11 GHz……………………………. 61

Fig. 2.20. The patch antenna designed for center frequency of 24 GHz………………. 62

Fig. 2.21. The radiation pattern of the patch antenna at 24 GHz…………………….. 63

Fig. 2.22. Return loss of the antenna at 24 GHz………………………………………. 64

Fig. 2.23. The photograph of the IF circuit……………………………………………… 65

Fig. 2.24. The measured array factors of the IF-network compared to the simulation. 66

Fig. 2.25. The photograph of the LO circuit and the patch-antenna elements together
with their feed lines…………………………………………………………… 67

Fig. 2.26. The phased array radiation pattern measurement setup……………………. 68

Fig. 2.27. The measured array patterns at 24 GHz compared to the array factors
calculated using the simulated S-parameters………………………………….. 69

Fig. 2.28. Measured input reflection coefficient of the IF circuit versus frequency at
different scan angles…………………………………………………………… 70

Fig. 2.29. The measured array factor at 2 GHz versus frequency……………………. 71

Fig 3.1. The complexity of varactor loaded transmission line is determined by the
amount of phase shift required from the phase shifter……………………… 75

Fig 3.2. Phased arrays using parallel feed network require maximum phase shift of (N-
1)θ from phase shifters to achieve inter-element phase shift of θ…………… 76

Fig 3.4. Series-fed phased array with phase shifters inserted in parallel require maximum
phase shift of θ from phase shifters to achieve inter-element phase shift of θ. 76

Fig 3.4. Series-fed phased array with phase shifters inserted in series require maximum

vii 
 
 

phase shift of θ from phase shifters to achieve inter-element phase shift of θ. 76

Fig 3.5. Bi-directional series-fed phased arrays require maximum phase shift of θ/2 from
phase shifters to achieve inter-element phase shift of θ……………………….. 77

Fig 3.6. The new bidirectional phased array utilizes directional couplers, switches and
phase shifters with a single control voltage to achieve beam steering………… 79

Fig 3.7. Array directivity versus the coupling factor for different array sizes…………. 81

Fig 3.8. Increase in array beamwidth compared to a uniformly excited array versus the
coupling factor for different array sizes…………………………………….. 81

Fig 3.9. The percentage of input power dissipated within the matched SPDT switch
versus the coupling factor for different array sizes…………………………. 83

Fig 3.10. Directivity of an eight-element series-fed array compared to an eight-element

uniformly excited array at (a) scan angle of zero and (b) scan angle of thirty
degrees………………………………………………………………………… 84

Fig 3.11. Amplifier can be used in the receive phased array to achieve a particular
amplitude tapering…………………………………………………………….. 85

Fig 3.12. Amplifier can be used in the transmit phased array to achieve a particular
amplitude tapering……………………………………………………………… 85

Fig 3.13. A photograph and general diagram of loaded transmission line phase
shifter…………………………………………………………………….. 88

Fig 3.14. A photograph and general diagram of reflective type phase shifter……. 89

Fig 3.15. A photograph and general diagram of vector modulated phase shifter… 89

Fig 3.16. A photograph and general diagram of switch line phase shifter………… 90

Fig 3.17. The circuit diagram of the phase shifter…………………………………….. 91

Fig 3.18. The required series inductance versus the varactor capacitance to maintain the
same voltage amplitude at the input and the output node………………….. 92

Fig 3.19. Maximum insertion loss versus maximum achievable phase shift when a fixed
reactance value is used……………………………………………………….. 93

Fig 3.20. Simulated phase shift through the phase shifter…………………………….. 94

Fig 3.21. Simulated insertion loss through the phase shifter…………………………… 95

viii 
 
 

Fig 3.22. Simulated return loss of the phase shifter…………………………………… 95

Fig 3.23. A single voltage applied to the bias network at the input controls the phase
shift across the entire phased array……………………………………………. 96

Fig 3.24. A Photograph of the eight-element bidirectional phased array operating at 2.4
GHz…………………………………………………………………………… 97

Fig 3.25. Measured phase shift through the phase shifter versus frequency for different
control voltages……………………………………………………………….. 98

Fig 3.26. Measured insertion loss through the phase shifter versus frequency for different
control voltages……………………………………………………………….. 99

Fig 3.27. Measured return loss of the phase shifter versus frequency for different control
voltages………………………………………………………………………… 99

Fig 3.28. Comparison between the measured and simulated array factors for the eight-
element phased array…………………………………………………………. 100

Fig 3.29. Input return-loss for different bias voltages………………………………… 101

Fig. 4.1. Conventional design of phased arrays requires a separate phase shifter for each
antenna………………………………………………………………………… 106

Fig. 4.2. The proposed phased array design requires a single phase shifter placed at the
opposite end of a serially excited antenna array to the input………………. 106

Fig. 4.3. The block diagram of phased arrays based on the proposed approach requiring a
single phase shifter throughout the entire phased array…………………….. 108

Fig. 4.4. Phase progression in the entire array can be controlled by tuning the phase of
one component of the signals at the antenna elements……………………… 108

Fig. 4.5. The block diagram of phased arrays based on the proposed approach requiring a
single phase shifter throughout the entire phased array……………………… 109

Fig. 4.6. The first signal components are achieved as the signal is traveling away from
input……………………………………………………………………………. 111

Fig. 4.7. The second signal components are achieved as the phase shifted signal is
reflected back into the feed network…………………………………………. 111

Fig. 4.8. The overall signal the each antenna element is achieved by combining two
signal components…………………………………………………………….. 112

Fig. 4.9. Approximately linear phase progressing can be achieved across the array… 112

Fig. 4.10. The amplitude variation at the antenna elements as the phase progression is

ix 
 
 

tuned…………………………………………………………………………… 114

Fig. 4.11. Circuit diagram of the phase shifter………………………………………….. 115

Fig. 4.12. Phase shift of the signal reflected by the phase shifter………………………… 116

Fig. 4.13. Loss of the signal reflected by the phase shifter………………………………. 116

Fig. 4.14. Two sub-array are connected to form a symmetrical structure fed from the
center…………………………………………………………………………… 117

Fig. 4.15. Photograph of the symmetrical phased array……………………………….. 120

Fig. 4.16. Measured array factor compared to the simulation…………………………... 120

x 
 
 

List of Tables

Table. 1.1. Comparison between the presented extended resonances phased array and the
published series-fed phased arrays…………….…………………………..……. 72
Table. 2.1. Comparison between the bi-directional series-fed phased array presented in
this chapter and the published series-fed phased arrays…………………………
103
 

xi 
 
 

ABSTRACT

This dissertation presents three new approaches to the design of phased array

antennas in order to reduce their complexity. The first approach is based on extended

resonance technique which, unlike conventional phased array designs, achieves power

dividing and phase shifting tasks within a single circuit. A new extended resonance

circuit is developed here that increases the maximum achievable scan angle by three

times compared to the extended resonance phased array demonstrated previously. In

order to expand the size of phased array, a new modular approach is used enabling a

scalable design of extended resonance phased array for the first time. By applying

heterodyne-mixing concept, a modular 24 GHz phased array has been demonstrated.

The second approach presented in this dissertation is based on a bi-directional

feeding method. A new phased array is designed based on this approach which demands

less phase shift from phase shifters compared to any of common phased arrays. The new

bi-directional phased array allows for beam steering using only a single control voltage.

A general design procedure for a bidirectional N-element phased array feed network is

presented for the first time which allows applying this approach to phased arrays with any

number of antenna elements. Furthermore, a new, compact phase shifter is designed and

xii 
 
 

utilized in the phased array. A 2.4 GHz bi-directional phased array has been designed and

fabricated.

Finally, the third approach described in the dissertation allows the phase

progression across the antenna elements to be controlled by using a single phase shifter.

Therefore, the number of phase shifters required in the phased array is substantially

reduced compared to conventional phased array designs which require a separate phase

shifter per each antenna element. A variable phase shift is achieved in this approach by

vector summation of signals. The amplitude ratios of these vectors are adjusted to provide

a linear phase progression. This approach is much simpler than the traditional Cartesian

phase shifting scheme. A 2 GHz phased array designed based on this approach has been

fabricated and tested.

xiii 
 
 Chapter 1

Introduction

Phased array antenna is a multiple-antenna system in which the radiation pattern

can be reinforced in a particular direction and suppressed in undesired directions. The

direction of phased array radiation can be electronically steered obviating the need for

any mechanical rotation. These unique capabilities have found phased arrays a broad

range of applications since the advent of this technology.

Phased arrays have been traditionally used in military applications for several

decades. Recent growth in civilian radar-based sensors and communication systems has

drawn increasing interest in utilizing phased array technology for commercial

applications. Many applications that can potentially benefit from phased arrays will be

discussed in this chapter.

Despite the broad range of potential phased array applications, phased array

technology has not been widely deployed in the commercial arena. The high cost and

complexity of phased arrays is the primary impediment to their deployment in any large-

scale commercial application. Thus, a substantial reduction in the cost and complexity of

1
phased array systems will facilitate their much wider use. In order to deploy phased

arrays in a wide range of applications, ongoing efforts are underway to develop new

architectures that allow complexity of phased array to be reduced. In addition to

conventional designs of phased arrays, various approaches proposed to realize phased

arrays with reduced complexity are discussed in this chapter. Finally, an overview of

thesis presenting several new approaches to design phased arrays is given at the end of

the chapter.

1.1 Background

Enhancing antenna directivity in order to improve long-range communication has

been subject of extensive research since the time of Marconi’s paper entitled “Directive

Antenna” [1]. The first demonstration of directive wireless communication which can be

considered as the origin of phased array technology was shown by Karl Ferdinand Braun.

He won the Nobel Prize for demonstrating enhanced transmission of radio waves in one

direction [2]. Up to 1920s, fixed beam antennas were primarily used to meet the

demands for increased gain. For instance, Beverage [3] invented so-called “wave

antenna” or “Beverage” antenna in order to increase the gain above that of dipoles and

other simple elements. The Beverage antenna was basically a long horizontal wire of

about one wavelength mounted over earth, and loaded with appropriate resistances.

Many forms of directive antenna array with high gain and resolution had been

developed by mid 1930s [4]. However, in addition to increasing directivity, there was

another issue which had not been addressed. The problem lied in the fact the received

signal did not arrive from one elevation angle, but from waves which arrive at different

2
Fig. 1.1. Multiple unit steerable antenna array developed by Friis and Feldman for short range
reception [4]

vertical angles that could change with time. Therefore, the signal would arrive outside of

the angular range of the narrow band antenna much of the time.

In order to track the receive signal as its directions changes, a multiple unit

steerable antenna was developed in 1937 by Friis and Feldman for short wave reception

(Fig. 1). Basically they used mechanical phase shifters to provide the required phase

progression across the antenna array. This early demonstration of electromechanical

scanning certainly opened the door to many years of technology development in order to

realize fully electronic scanning systems.

3
 Although scanning antenna arrays had been developed for various electronic

communications applications, radar system was the first application for which phased

array was broadly used for the first time. Radars demanded extremely rapidly scanning

beams, wide band performance, the ability to adapt the beam patterns to avoid jammers

and clutter which could be addressed by phased array antennas. Friis and Lewis [5]

developed an electromechanical phased array for radar application using rotary phase

shifter developed by Fox [6]. The array was center-fed and used extra fixed line lengths

to equalize phase across the array with all phase shifters set to zero. The phase shifters

were ganged together and scanned the beam continuously with inter-element progression

of 3600◦/s. During World War II, phased arrays were integrated into the radar systems by

another Nobel Prize winner, Luis Alvarez, to aid in the landing of airplanes in England

[7].

Application of phased array technology later extended beyond military arena and

brought another Nobel Prize for its users, Antony Hewish and Martin Ryle, who had

adapted phased arrays for radio astronomy applications [8]. Nowadays, beside the

established and sophisticated military applications, phased arrays are used for a wide

range of commercial applications as well. Some of these applications are discussed in the

next section.

1.2 Emerging applications of Phased array antennas

Phased arrays have been traditionally used for military radar applications for

many decades. Radar systems either based on transmitting continuous wave or pulsed

signal, emit a signal at a particular direction and collect the reflected signal in order to

4
 Fig. 1.2. The Sea based X-band radar is a combination of the world’s largest phased array radar
carried aboard a mobile, ocean going semi-submersible platform

process, extract the information and identify the object. Phased array radars allow a

warship to use one radar system for surface detection and tracking, air detection and

tracking (finding aircraft and missiles) and missile uplink capabilities [9]. The electronic

beam steering of the phased array radar is orders of magnitude faster than the traditional

radars based on mechanical rotation. The largest phased array in the world is a $900

million dollar radar operated by US Missile Defense Agency. (Fig. 1.2)

Another important application that can take the advantage of phased arrays is

automotive collision avoidance radar or adaptive cruise control technology [10]. As

5
 Fig. 1.3. Using radar sensors in the car can provide the driver with several driving-aid

functions

shown in Fig 1.3, using radar sensors in the car can provide the driver with several

driving-aid functions such as automatic cruise control (ACC), parking aid, blind spot

detection, and side collision warning [11]. High resolution radar systems having powerful

signal processors capable of advanced image processing can further enable object

classification, roadside detection, and prediction of the lane course, therefore enabling

intelligent interpretation of traffic scenes [12]. Finally, autonomous driving can be

realized by combining short-range radar, global positioning techniques, and wireless

communications. Every minute, on average, car accidents take life of at least one person

[13]. This results in more than 42,000 deaths and 3,000,000 injuries a year in the United

States alone. The financial damages caused by car accidents can be as high as 1–3 percent

of the world’s gross domestic product every year. For the United States, this exceeds

$200 billion a year. Collision avoidance radars can reduce these staggering numbers by

6
monitoring an automobile’s motion and its surroundings in order to warn and prevent car

accidents by activating the breaks in threatening situations. The current collision

avoidance radars can cost as high as $1,500 to $3,000. Due to their high cost, the radar is

an option only on luxury models, such as the Mercedes S-class and Jaguar XKE.

Developing low cost phased-arrays operating at frequency bands allocated by Federal

Communications Commission (FCC) that can be deployed in ordinary cars is currently

subject of extensive research.

It is believed that the emerging gigabit wireless personal communication systems

at 60 GHz will be the first product with a large commercial market in the mm-wave

frequency band. Phased arrays can play a significant role in success of these systems

(Fig. 1.4). Spatial selectivity of phased array beams can increase the channel capacity of

the gigabit wireless personal communication systems to achieve data-rates that have not

been possible before [14-18]. Furthermore, phased arrays can facilitate the deployment of

these systems by significantly reducing their power consumption through more efficient

power management. In the United States, the emerging gigabit 60 GHz communication

systems are allowed to transmit up to 39 dBm equivalent-isotropically-radiated power.

Such high power levels are very challenging to generate using the low cost commercial

integrated circuit technologies. Exploiting phased arrays in these systems can alleviate

this problem by providing beam forming and increasing the radiation gain in a desired

direction [14, 17, 18]. This will obviate the need for prohibitively large output power

levels from millimeter-wave power amplifiers. Finally, by using phased arrays, more

reliable and robust communication links can be achieved through mitigating the problem

of multipath fading and co-channel interference and by suppressing the signals emanating

7
 Fig. 1.4. SiBEAM 60 GHz phased array transceiver for wireless video-area network

from undesirable directions. The integrated circuit technologies now show a promising

performance at mm-wave frequencies. Low cost, high integration density of the existing

technologies allows for a system-on-chip implementation of mm-wave systems, and is

expected to provide an unprecedented opportunity for application of phased array

technology in the next generation of communication systems.

Phased arrays are widely used in base stations to enhance the signal coverage in a

particular area while minimizing interference to other areas [19]. Satellite TV systems

available in the current commercial market also employ phased arrays. Compared to

traditional parabolic dish systems, phased array based broadcasting-reception links have

the advantages of being more robust in inclement weather, having smaller profile and

lower weight that allows them to be more easily mounted on walls and roofs. Moreover,

8
 Fig. 1.5. HAARP is a research station directed by the Air Force Research Laboratory's Space
Vehicles Directorate in Gokona, Alaska to gather data about the atmosphere

the adaptive beamforming allows mobile objects such as planes and vehicles to have

access to satellite programs [20].

Biomedical applications are among the emerging applications that can take the

advantage of phased-array [21-23]. For instance, phased array has been used in

microwave imaging to detect early stage breast cancer [24-26]. In this method, a

wideband impulses are emitted by an antenna array placed at the breast surface. The

receiver uses beam-forming to focus the backscattered signal from the malignant tumor

and compensate for the frequency dependent propagation effect. The malignant tumor

9
and breast tissue causes a dielectric discontinuity that affects the signal reflection. The

microwave imaging can provide much higher detection probability than X-rays or

ultrasound as the relevant contrast is an order of magnitude higher for microwave. The

cost of microwave imaging is also much lower than other current alternatives such as

magnetic resonance imaging (MRI) and is less harmful to the patients than X-ray.

1.3 Phased Array Principle

The block diagram of an N-element phased array is shown in Fig. 1.6. “N”

identical antennas are equally spaced by a distance “d” along an axis. For simplicity, we

discuss only the receiver case, while similar concepts are held for transmitter as well due

to the reciprocity concepts. Separate variable time delays are incorporated at each signal

path to control the phases of the signals before combining all the signals together at the

output. A plane-wave beam is assumed to be incident upon the antenna array at an angle

of θ to the normal direction. Because of the spacing between the antennas, the beam will

experience a time delay equal to Eqn. 1.1 in reaching successive antennas.

2πd sin( θ )
Δτ = (1.1)
λ

Here, λ is the wavelength of the signals.

Hence, if the incident beam is a sinusoid at frequency ω with amplitude of A, the

signals received by each of the antennas can be written as Eqn. 1.2.

Si = Ae− jnΔτ (1.2

10
 Fig. 1.6. Block diagram of N-element phased array

The plane wave incident at an angle upon the phased array experiences a linear

delay progression at the successive antenna elements. Therefore, the variable delay

circuits must be set to a similar but with reverse delay progression to compensate for the

delay of the signal arrived at the antenna elements. In linear arrays, variable time delays

are designed to provide uniform phase progression across the array. Therefore, the signal

in each channel at the output of the variable delay block can be written as Eqn. 1.3.

11
 S ' i = Ae − jnΔτ e − jnα (1.3)

In this equation, α denotes the difference in phase shift provided by two

successive variable time delay blocks. Therefore, array factor which is equal to the sum

of all the signals normalized to the signal at one path can be written as Eqn. 1.4.

N
F= ∑ e − jn( Δτ −α ) (1.4)
n =1

According to Eqn. 1.4, the peak of the array factor occurs at an incident angle

which can be determined by Eqn. 1.5.

2πd α
sin( θ ) = α Or θ = arcsin( ) (1.5)
λ 2πd

At this incident angle, which is called scan angle, the linear delay progression

experienced by the wave arriving at the successive antennas is perfectly compensated

with the time delays incorporated at each path resulting in all the signals being coherently

combined at the output of the receive array. The array factor can also be shown as Eqn.

1.6.

N 2πd
sin 2 [ ( sin( θin ) − α )]
F= 2 λ (1.6)
2πd
sin 2 [ sin( θin ) − α ]
2λ

At the scan angle “θ”, the array factor has a maximum value of N2. For other

angles of incident, the array factor will be lower than this value indicating spatial

12
selectivity of phased array. In addition to the spatial filtering, one of the main capabilities

of phased array is that the peak gain of the array, according to Eqn. 1.5, can be controlled

by electronically tuning the variable time delays eliminating the need for any mechanical

rotation of antenna array.

It should be also noted that the benefits of using phased array is increased as the

number of elements in the phased array is increased. As mentioned before, the maximum

value of array factor N2 is directly proportional to the number of array elements.

Furthermore, the beam width of the array can be decreased by increasing the number of

array elements in order to enhance the spatial selectivity of phased array.

1.4 Advantages of using phased array

1.4.1 Spatial filtering

A major advantage of incorporating phased array into communication systems is

its capability of spatial filtering. Phased arrays are able to suppress the signals emanating

from undesired directions; even if omnidirectional rating elements are used. In general,

the ultimate radiation/reception pattern of a phased array is determined by multiplying the

received pattern of a single antenna element by the array factor, assuming identical

13
 Fig. 1.7. Array factor for a linear 4-element phased array

antenna elements are used across the entire array. In order to demonstrate the spatial

filtering capability of phased arrays, the array factor for a linear 4-element phased array is

shown in Fig. 1.7 assuming all the antenna elements are identical and placed λ/2 apart,

where λ is the wavelength. As can be seen, the array radiates/receives signals with a

beamwidth 26 degrees centered at 30 degrees while the undesired signals emanating from

other angles are significantly attenuated. As mentioned before, the ultimate radiation/

reception pattern of a phased array is determined by multiplication of a pattern of each

individual antenna with the array factor. Therefore, by different arrangement of antenna

elements, various radiation/reception patterns can be achieved to fulfill the required

spatial filtering. For instance, array factors for linear 4-elemet phased array having

14
 10

5
Array_factor

-5

-10

-15

-20
-100 -75 -50 -25 0 25 50 75 100
180*TX/pi

Fig. 1.8. Array factor for 4-elemenet array with different antenna spacings

different spacings, λ/1.5, λ/2 and λ/3 are compared in Fig. 1.8. As can be seen, by

increasing the spacing between the antenna elements the beamwidth of the array radiation

pattern decreases. However, increasing the spacing can cause in increase of side lobe

level and emergence of grating lobes. Furthermore, in phased-array systems the

amplitude excitation of the array elements can be tapered to lower the side lobe levels or

to adjust the null positions at directions where the strong interfering signals are

emanating from [27].

It should be noted that antenna elements can be placed in two dimensions as well

to create spatial filtering in azimuth and elevation. Furthermore, the arrangement of the

15
antenna elements and their feeds can be changed in conjunction with the associated

circuitry to allow for the multiple beams to be created and steered independently.

Spatial filtering of phased arrays not only allows for suppressing the interfering

signals emanating from undesired sources, but, it also can alleviate the problem of fading

and multi-channel interference [28]. Fading is caused when the signal travels along

multiple paths before arriving at the receive antenna. Therefore, although all of these

signals have been emanated from a single transmitter, the phases of these signals arriving

at the receive antenna are different and can even cancel out each other. Therefore, it is

desired to receive the signal from a specific path while suppressing the others

propagating along other directions. Spatial filtering of phased array can be exploited to

alleviate this problem.

1.4.2 Gain boosting in transmit phased array

In general, the maximum distance that reliable communication between a

transmitting and receiving system can occur is determined by the maximum power

transmitted in the direction of receiver. For a given receiver sensitivity, the power

transmitted in the direction of receiver can significantly affect the maximum range of

communication. In radars, also, the maximum transmit power determines the maximum

range that radar would be able to detect a target with a given radar cross section.

16
 Parray = N ∑ Ppath

Fig. 1.9. Phased array transmitter can boost the gain in a desired direction

By implementing transmitter antennas using phased arrays one can boost the

transmit power level in a desired direction while the total power transmitted remains the

same. Therefore, compared to omnidirectional transmitter, phased array transmitter are

capable of communicating over longer distance for a given receiver sensitivity (Fig. 1.9).

An omnidirectional transmitter radiates electromagnetic power in all directions, and only

a small fraction reaches the intended receiver. Not only is a major fraction of this power

wasted, but it also causes interference for other users. With ever-increasing wireless

applications and the rapid growth of number of wireless users, data-rates currently

17
 Fig. 1.10. Phased array receiver can improve the noise performance of the system

achievable in available wireless communication networks are in general limited by the

interference rather than noise. An increase in transmit power for all users to increase the

achievable range and data rate increases the interference level as well and therefore does

not provide any benefit to the system overall capacity.

1.4.3 Sensitivity improvement in receive phased array

Sensitivity of a receiver can be directly obtained from the signal to noise ratio at

the output of the receiver’s front end. In general, a signal to noise ratio above a certain

level is required to detect the signal with a target error probability in mind. The minimum

signal to noise ratio possible would determine the minimum detectable signal at the

18
antenna element. Therefore, the noise figure of the receive systems is improved and the

overall sensitivity of receive systems is enhanced. Having the noise figure NFchannel and

minimum acceptable signal to noise ratio at the output SNRout, the signal to noise ratio at

the input of a single path receiver is given in Eqn. 1.7.

SNR In ( dB ) = SNRout ( dB ) + NFchannel (1.7)

In a receiver based on using phased array, the signals at the output of phased array

are combined coherently for transmit signal coming from the desired direction. Assuming

the current from the antenna elements are combined coherently, the current at the output

will be N times larger than the current at each individual antenna; therefore, the signal

power level at the output will be N2 times larger than the received signal.

S out = N 2 S in (1.8)

Here, n is the number of antenna elements

The noise contributed by the receive channels are not coherent, therefore, at the

output of the phased array, the power of noise from different channels are combined.

Output noise level, therefore, will be N times larger than the noise contribution of each

channel N. As a result, the noise figure of the entire system can be calculated as shown in

Eqn. 1.9.

nPN P NFchannel
NFarray = =n N = (1.9)
n 2 PS PS N

19
 Therefore, the minimum acceptable SNR at the input of phased array can be

shown as Eqn. 1.10. Assuming the noise performance of the channels remains the same

as when used alone.

NFchannel
SNR In ( dB ) = SNRout ( dB ) + (1.10)
N

Thus, compared to the output SNR of a single-path receiver, the output SNR of

the array is improved. This improvement can be more significant if number of antenna

elements is increased. In general, the improvement in sensitivity of receiver by using

phased array is given in Eqn. 1.11. For instance, a 16-path phased-array can improve

receiver sensitivity by 12dB.

N
Sensitivity array ( dB ) − Sensitivity channel ( dB ) = 10 log10 (1.11)

1.5 Phased Array Architecture

1.5.1 Phased Arrays based on feed network design

Phased arrays are usually composed of a feed network and a number of phase

shifters. Feed networks are used to distribute the output signal of the transmitter to the

radiation elements and phase shifters control the phase of the signals at each radiating

element to form a beam at the desired direction. There are almost as many ways to feed

arrays as there are arrays in existence. In general, array feed networks can be classified

into three basic categories: constrained feed, space feed and semi constrained feed which

is a hybrid of the constrained and the space feeds [29]. In a space feed network, the array

20
 Fig. 1.11. General diagram of parallel fed phased array

is usually illuminated by a separate feed horn located at an appropriate distance from the

array [30]. Due to the free space existing between the feed and radiating elements, this

type of feed network is not a good candidate for planar arrays. The constrained feed,

which is usually the simplest method of feeding an array, generally consists of a network

which takes the power from a source and distributes it to the antenna elements with a feed

line and passive devices. The constrained feed itself can be categorized into two basic

types: parallel feeds and series feeds [31]. The architectures based on these two types of

feed network are the most common approach to design phased arrays.

1.5.1.1 Parallel-fed Arrays

In parallel feed networks, which are often called corporate feeds, the input signal

is divided in a corporate tree network to all the antenna elements as shown in Fig. 1.11.

These networks typically employ only power dividers [32]. Therefore their performance

critically depends on the architecture of the power splitter/combiner used. 2N number of

21
 Fig. 1.12. General diagram of series fed phased array

radiation elements is preferred for these types of arrays, where N is the number antennas

in the phased array.

1.5.1.2 Series-fed Arrays

In a series-fed array the input signal, fed from one end of the feed network, is

coupled serially to the antenna elements as shown in Fig. 1.12. The compact feed

network of series-fed antenna arrays is one of the main advantages that make them more

attractive than their parallel-fed counterparts. Beside compactness, the small size of

22
series-fed arrays results in less insertion and radiation losses by the feed network [33].

The cumulative nature of the phase shift in series arrays also relaxes the design

constraints on the phase tuning range of the phase shifters. In an N-element series-fed

array, the required amount of phased shift is smaller than parallel fed arrays by a factor of

(N-1). However, the cumulative nature of phase shift through the feed network results in

an increased beam squint versus frequency [34], which is one of the main limitations in

series-fed designs. The loss through the phase shifters is also cumulative in series fed

arrays which can be an issue in the design of arrays with a large number of array

elements.

1.5.2 Phased array based on phase shift stage

Phase shifters can be placed at any stage of phased array. Based on the stage in

which phase shifting is performed, phased arrays can be categorized into four district

types: RF-phase shifting, LO-phase shifting, IF-phase shifting and digital phased arrays.

In the following sections, each of these types of phase arrays is discussed.

1.5.2.1 RF phase shifting

Fig 1.13 displays the general architecture of phased arrays using RF phase

shifting. In this architecture, the signals at the antenna elements are phase-shifted and

combined in the RF domain. The combined signal is then down converted to baseband

using heterodyne or homodyne mixing.

Designing phased arrays using RF phase-shifting has been traditionally more

common compared to the other architectures. Since this technique requires only a single

23
 Fig. 1.13. General diagram of RF phase shifting phased array

mixer and there is no need of LO signal distribution, it usually results in the most

compact architecture among other phase array designs [35-38]. Furthermore, it provides

insulation of a larger portion of the receiver chain from interference emanating from

undesired directions. Moreover, since phase shifting and the signal combining are

performed prior to down conversion in the RF phase shifting architecture, the interferer is

filtered at RF stage prior to the mixer; therefore, the requirement on dynamic range of

down conversion mixer is not as stringent as other types of phased array architectures

[39].

24
 The main challenge of using phase shifting at RF stage in the design of phased

arrays is implementing high performance phase-shifters capable of operating at RF

frequencies. Regardless of the technology used to implement phase shifters such as BST,

MEMS, GaAs or CMOS, passive phase shifters, in general, tend to be excessively lossy

at microwave and millimeter regime [40, 41] while active phase shifters usually suffer

from low dynamic range [42, 43]. Dynamic range of phase shifters is particularly

important in the operation of phased arrays as phase shifters are required to operate in the

presence of strong interferers. Furthermore, phase shifter used in receive phase arrays are

in the RF signal path and therefore, their noise performance can be critical for the

sensitivity of receivers. Another factor that should be taken into consideration is the

amplitude variation of the signal at each channel. As the signal combining and null

forming at undesired directions is significantly affected by the amplitude of the signal at

each channel, the phase shifters are required to not just have a low insertion loss but also

maintain a constant loss within their phase tuning range. Variable gain amplifiers can be

used to main the signal amplitude at each channel, however their insertion phase should

be constant as their amplitudes are changing.

1.5.2.2 LO phase shifting

Fig 1.14 shows the general architecture of phased arrays based on LO phase

shifting. The phase of RF signal at each channel is basically the sum of phases of IF and

LO signals. Therefore, tuning the phase of LO signals would translate into changing the

phase of RF signals. The advantage of this approach compared to other architectures is

25
 Fig. 1.14. General diagram of LO phase shifting phased array

that the phase-shifters are not placed on the signal path [44, 45]. As a result, the loss,

nonlinearity and the noise performance of the phase-shifters would not have a direct

impact on the overall system performance. Furthermore, performance of the required

phase shifter on LO signal path, such as bandwidth, linearity and noise figure will not be

as stringent as the phase shifters on the signal path [46]. However, this method compared

to RF phase shifting requires a large number of mixers, therefore; in general, the overall

complexity and power consumption will be higher than phased arrays based on RF stage

phase shifting.

26
 LNA

Mixer

Phase
shifter

IF Power Combiner IF

LO Power Divider LO

Fig. 1.15. General diagram of IF phase shifting phased array

1.5.2.3 IF phase shifting

The general architecture of phased arrays based on IF phase shifting is shown in

Fig. 1.15. As mentioned before, the phase of RF signal at each channel is the sum of

phases of IF and LO signals. Therefore, tuning the phase of RF signals can be achieved

by tuning through tuning the phase of IF signal [47, 48]. The main advantage of this

approach over RF and LO phase shifting is that the phase-shifting is performed at much

lower frequencies, therefore, designing phase shifters with much better performance can

be possible at IF path. As a result, the loss, nonlinearity and the noise performance of the

phase-shifters can be much better when IF phase shifting is used. However, in this

architecture similar to LO phase shifting phased arrays large number of mixers are

27
required that can add to the overall system complexity and its power consumption.

Furthermore, since the interfere cancellation occurs only after the IF stage, all the mixers

are required to have a high level of linearity capable of handling strong interference

emanating from undesired directions.

1.5.2.4 Digital Phased Arrays

In digital phased array design, the signal at each array element is digitized using

an Analog-to-Digital Converter (ADC) and the digital signal are then processed using a

Digital Signal Processing unit (DSP) [49]. In this architecture, interference emanating

from undesired directions is not canceled out after signal processing, therefore, all the

elements including the RF mixer and ADC and the DSP unit must have sufficient

dynamic range capable of handling the interferers. Furthermore, in general each channel

requires the entire RF chain front end circuits. As a result, power consumption is

relatively high in this architecture.

The main advantage of the digital array is its multifunction capability such as

creating many beams [50]. An extensive variety of complex, signal-processing algorithms

can be implemented using DSP units. For instance, multi-beam and multiple-input-

multiple-output (MIMO) functionality can be achieved by the digital phased arrays. Such

phased arrays can be capable of distinguishing among desired signals, multipath and

interfering signals, as well as demonstrating their directions of arrival. Furthermore,

digital phased arrays can adaptively update their beam patterns, so as to track the desired

28
 RF Input RF Output

Fig. 1.16. Photograph and boding diagram for a 5-bit phase shifter

signal with the beam’s main lobe and track the interferers by placing nulls in their

directions.

1.6 Challenges in design of Phased Array Antennas

As discussed in the last sections, there are numerous areas that can benefit from

phased arrays. However, there are disadvantages associated with the current

implementations of phased arrays that can deter their broad deployment. In order to pave

the way for wider employment of phased arrays in various applications, there are several

challenges that should be addressed. The primary limitations of utilizing phased arrays

consist of cost, size, weight, power consumption, and their high complexity. These

constraints have restricted this technology to military, aerospace arena which usually

require low quantity prototypes. The systems used in this arena usually can afford to

29
accommodate high cost and complexity of phased arrays in return of high performance to

meet their stringent requirement. It is expected that by addressing these technical hurdles

and substantially reducing system costs, more ubiquitous use of phased arrays

particularly in commercial markets will be possible [51]. Thus, any substantial reduction

in the cost and complexity of phased array systems will facilitate their much wider

consumer use which requires unit costs well below what is currently possible.

The high cost and complexity of phased array typically ties in with the cost of

multiple receive or transmit/receive modules, whose output and control data are

multiplexed into a backend processor unit, which is separate from the T/R modules. One

of the critical elements of these T/R modules is a phase shifter. Phased arrays usually

require complex integration of many expensive solid-state, MEMS or ferrite based phase

shifters, control lines, together with power distribution networks. Phase shifters typically

contribute to a significant amount of the cost of producing a phased array antenna and it

is not uncommon for the cost of the phase shifters to represent nearly half of the cost of

the entire phased arrays [52]. Furthermore, a major complexity of phased array design

can be attributed to the complexity of their phase shifters. Since the advent of the array

theory and development of the early beam steerable phased arrays, phase shifters have

been widely recognized as the most complex and sensitive part of a phased array. The

complexities in the corporate feed network and biasing network for the phase shifters as

well as their parasitic radiation and interactions with the radiating elements, render

implementation of phased arrays very challenging. This is particularly true where a large

number of phase shifters with accurate phase control are required.

30
 Conventional linear phased arrays, capable of scanning in azimuth or elevation,

require one phase shifter circuit for each antenna element. In two dimensional scanned

arrays, the number of required phased shifters is even larger. Each phase shifting circuit

is connected to bias lines and the associated driver circuitry for its phase control. To

better illustrate the level of complexity introduced by the individual phase shifters within

the array, a photograph of a five-bit digital phase shifter by Triquint along with the

required bias and control connections is shown in Fig. 1.16. Independent of the type of

phase shifter used, design of phased array systems is considered to be complicated and

costly mostly due to the number and cost of phase shifting elements required. Thus, in

order to reduce the cost of phased arrays, ongoing efforts are underway to develop new

architectures that allow for designing phased arrays with reduced complexity. In the next

section, different approaches proposed by researchers to allow for phased array cost and

complexity reduction are discussed in more details.

1.7 Review of state of Art Phased array Design

A common approach to reduce the required number of phase shifters is based on

grouping the radiating elements into a number of sub-arrays each using a single phase

shifter [53-55]. The general diagram of phased arrays designed based on this approach is

shown in Fig. 1.17. In general, the concept is based on replacing the linear-phase profile

of the array excitation by its coarse staircase approximation in this type of phased array.

As can be seen in Fig. 1.17, the array elements are divided into the groups of in-phase

elements, or sub-arrays, referred to as a primary array. Each of the primary arrays

assuming to be identical uses a single phase shifter. Furthermore, each of these primary

31
 Feed Network Feed Network Feed Network

Feed Network

Fig. 1.17. Grouping antennas into a sub array each using one phase shifter can reduce the
number of required phase shifters

arrays can be viewed as the elements of a second phased array called the secondary array.

The array factor for the overall structure will be equal to the product of these two array

factors. The main problem with this approach is increased side-lobes and grating lobes

limiting the array scan range [56-58].

Another beam steering technique [59-61] is based on tuning an array of parasitic

radiators to replace phase shifters with individual varactors. As shown in Fig. 1.18, in

this approach, the sole RF port is connected to one central antenna element and a number

of surrounding parasitic antenna elements are terminated with tunable reactances. Beam

steering is achieved by tuning the load reactances at parasitic elements surrounding the

central active element. The load reactances are usually realized by the reversely biased

varactor diodes. One of the issue in this approach is it limited directivity. Since the

beamforming in this approach relies on interaction of parasitic and main radiators,

32
 Fig. 1.18. Beam steering can be achieved by tuning the load reactances at parasitic elements
surrounding a central active element [51]

achieving narrow beams can be challenging as the level of radiation from parasitic

antenna elements is comparatively low to the radiation of main array element.

Implementing a planar array using this technique is another issue that has been subject of

research.

A phase shifter-less beam steering antenna array can also be realized by

exploiting frequency-controlled scanning [62-64]. However, most practical systems

require the radiating frequency to be kept fixed as the beam angle is steered. This can be

achieved by using heterodyne mixing concept in the frequency-scan array. As shown in

Fig. 1.19, in this approach, the phase shift between the antenna elements is usually

33
 Fig. 1.19. Frequency scanning can be utilized to achieve beam steering [54]

obtained by changing the frequency of either LO signal fed to frequency sensitive

network while the frequency of the IF signal is changed as well in order to maintain a

constant radiation frequency.

Another phased array design approach is presented in [65-67] to control the phase

progression in a phased array by detuning the peripheral elements in an array of free

running coupled oscillators. The principle of operation of coupled oscillator array is

based on the concept of injection-locking which was studied by Robert Adler for the first

time [68]. The underlying idea is based on the fact that when an electrical oscillator is

injected with an external signal whose frequency is close to the free-running frequency of

34
 Fig. 1.20. Phased array using coupled oscillator array [56]

the oscillator, the oscillator becomes synchronized in frequency to the external signal. As

shown in Fig. 1.20, the concept of injection locking can be extended to a linear array of

coupled oscillators. An array of free running oscillators are arranged in a linear format

and each oscillator is injected with the portion of the outputs of its nearest neighbors. As

the edge oscillator elements are detuned in frequency, the oscillator outputs become

synchronized to a common oscillation frequency and exhibit a linear phase progression.

1.8 Overview of Thesis

This thesis presents several new phased array design approaches that allow the

complexity of phased array to be reduced. In the second chapter, a new phased array

design based on extended resonance is presented. In this method, power dividing and

35
phase shifting tasks are performed using the same circuitry. Unlike the conventional

phased arrays, this technique eliminates the need for a separate phase shifter per each

antenna element hence reducing their complexity. This technique is utilized along with

heterodyne mixing to allow phase shifting to be performed at a lower frequency. In this

phased array design, extended resonance technique is utilized to achieve uniform power

distribution at the LO/IF paths and to perform phase-shifting at the IF-stage. A new

extended resonance circuit topology has been employed in the IF stage to enhance the

scan range of the phased array. Furthermore, comprehensive analysis of the maximum

achievable phase shift in the presented extended resonance circuit is given. Finally, a

complete modular 7-element 24 GHz phased array has been demonstrated. The

architecture of the proposed phased array is discussed and the measurement results are

presented as well to validate the design.

In the third chapter, different types of array architectures and their required

amounts of phase shift are discussed and a new bi-directional series fed phased array is

introduced which demands less phase shift from phase shifters compared to any of the

conventional designs of phased arrays. As the result, complexity of the phase shifters

required in the array can be reduced resulting in overall phased array cost and complexity

reduction. A general design procedure for a bidirectional N-element phased array feed

network is presented. Furthermore, a compact phase shifter is designed and utilized in the

phased array. The overall architecture allows one to steer the radiation beam by a single

control voltage obviating the need for complex bias circuits. As a proof-of-principle, an

eight-element bi-directionally fed phased array at 2.4 GHz is designed and fabricated.

36
The measurement results showing good agreement with the simulations are given in the

chapter.

Chapter four presented a new approach to design phased arrays. Unlike the

conventional designs of phased arrays requiring a separate phase shifter for each antenna

element, in the proposed design, the phase progression across the antenna elements can

be controlled by using only a single phase shifter. As a result, the cost and complexity of

the phased arrays can be significantly reduced based on this design. The proposed phased

array is composed of couplers, amplifying stages, power dividers and two phase shifters.

The phase shifters are placed at the two opposite ends of a serially fed array.

Furthermore, as a proof-of-principle a 2 GHz phased arrays is fabricated and tested. The

measurement results showing good agreement with the simulation are presented in the

chapter.

Finally, the last chapter concludes the thesis and outlines its main achievement.

Furthermore, direction for future work to continue the presented works is given.

37
 Chapter 2

A New Phased Array Design Based on Extended

Resonance Technique

2.1 Introduction

In this chapter, a new method to design phased arrays based on the extended

resonance technique is presented. In this method, power dividing and phase shifting tasks

are performed using the same circuitry. Unlike the conventional phased arrays (Fig. 2.1)

where there is a need for a separate phase shifter per each antenna element, this technique

eliminates this requirement hence reducing phased arrays complexity (Fig. 2.2). In the

design of the phased array, extended resonance technique is utilized with heterodyne

mixing to achieve uniform power distribution at the LO/IF paths and to perform phase-

shifting at the IF-stage. A new extended resonance circuit topology has been employed in

the IF stage to enhance the scan range of the phased array. In addition to the design

description of the IF-stage, comprehensive analysis of the maximum achievable phase

shift in the new extended resonance circuit is presented. Furthermore, a modular

approach to realize scalable phased array design is introduced. The design complexity

associated with large phased arrays can be reduced by using the scalable phased array
38
 Fig. 2.1. The conventional design of phased arrays requires one phase shifter for each antenna
element

Fig. 2.2. In an extended resonance phased array, phase shifters and power dividing network are
combined in a single entity

design introduced here. Finally, a complete modular 7-element 24 GHz phased array has

been demonstrated. The architecture of the proposed phased array is discussed in the next

39
sections, while the measurement results are presented in Section 2.6 to validate the

design.

2.2 Extended Resonance Phased Array Concept

The extended resonance technique was originally introduced to design power

combining oscillators using multiple solid-state devices [69-73]. The technique also has

an excellent feature of equally power combining and dividing. To explain the theory of

the extended resonance technique, we consider a circuit diagram of an extended

resonance module with N ports all terminated with the same conductance “G” as shown

in Fig. 2.3. Each series impedance Xn is chosen to transform the admittance seen at the

node to its conjugate resulting in voltages with equal magnitudes at all the nodes [52].

Vn +1
=1 (2.1)
Vn

However, a phase difference exists between these voltages, which directly depend

on the value of each shunt susceptances Yn. The shunt susceptances can be chosen

according to Eqn. 2.1 to achieve a linear phase progression at nodes V1 through Vn.

Yn = ( 2n − 1 )Y1 (2.2)

Therefore, the phase difference between each two adjacent nodes can determined

based on Eqn. 2.3.

Y
θ = 2 arctan( 1 ) (2.3)
G

40
 Fig. 2.3. The general structure of an extended resonance power dividing network

The required series admittance Xn to transform the admittance at the node n to its

conjugate is:

2Y1
Xn = (2.4)
n( G 2 + Y12 )

The shunt susceptance at the last node cancels out the imaginary admittance seen

at the node. Thus, the shunt admittance at the last node can be calculated as:

Y N = ( N − 1 )Y1 (2.5)

The admittance seen at the last node NG is matched to the source impedance

using a matching circuit.

In order to vary the relative phase of the output ports, tunable circuits can be

utilized to control the value of Yn. Furthermore, the series reactance Xn should also be

tuned simultaneously to maintain the equal power division. In the next section, a design

41
of a new tunable extended resonance circuit is presented. This circuit is optimized to

achieve a relatively large inter-element phase shift.

2.3 A new extended resonance circuit to enhance maximum achievable

inter-element phase shift

As mentioned before, in order to vary the relative phase of the output ports,

tunable circuits can be utilized to control the value of Yn and the series reactance Xn

simultaneously to maintain the equal power division. In general, if the shunt susceptance

Y1 can be tuned from jBmin to jBmax, the maximum achievable phase shift can be

determined by:

B B
Δθ = 2[arctan( max ) − arctan( min )] (2.6)
G G

Based on Eqn. 2.6, the maximum achievable phase shift depends not only on the

susceptance variation (jBmax-jBmin) but also on the susceptance value as well. It can be

shown that for a fixed susceptance variation- (jBmax-jBmin)- there is an optimum

susceptance that can result in the maximum achievable phase shift. This can be exploited

to increase the phase shift in the extended resonance circuit.

In the previous extended resonance circuits [74-76], Yn and Xn were replaced with

a series of varactor diodes as shown in Fig. 2.4. Therefore, the phase shift between power

divider ports when the varactors are tuned is:

ωC ωC
Δθ = 2[arctan( ) − arctan( )] (2.7)
G tG

42
 Fig. 2.4. The extended resonance circuit using just varactors as shunt susceptance

where t is the tunability of the varactor (the ratio of the maximum capacitance to the

minimum capacitance). Note that varactors at various ports have different values, but

they have the same tunability, t.

It can be shown that depending on the tunability of the varactor, there exists an

optimum normalized capacitive susceptance, which results in the maximum phase shift

between power divider ports, or maximum scan angle for the phased array [52].

To achieve the maximum phase shift it can be shown that value of capacitor

should be [52] :

wCopt
= t (2.8)
G

43
 Fig. 2.5. The proposed extended resonance circuit, with a block shown above, can considerably
increase the maximum achievable phase shift

The maximum phase shift then will be given by:

2 t
Δθ max = π − 2.tan−1{ } (2.9)
t −1

In order to extend the phase shift range beyond the limit given by Eqn. 2.9, the

modified design shown in Fig. 2.5 is presented. In this design, an inductor L1 is

incorporated in shunt with the varactor C1 to increase the achievable phase shift. In this

design, although the overall susceptance variation (jBmax-jBmin) remains the same, the

optimum inductance value can be determined to maximize the achievable phase shift. As

the shunt varactor C1 is tuned, the series reactance Xn should transform the admittance

seen at the node V1 to its conjugate in the entire tuning range. Such a reactance capable of

providing both inductive and capacitive impedance can be provided by placing a varactor

44
 300
Maximum Achievable Phase Shift (degree) Cmax=2 pF, t=2
250 Cmax=2 pF, t=4
Cmax=5 pF, t=2
Cmax=5 pF, t=4
200 Cmax=10 pF, t=2
Cmax=10 pF, t=4
150

100

50

0
0 1 2 3 4 5 6 7 8 9 10

Shunt Inductance(nH)

Fig. 2.6. Maximum achievable inter-element phase shift versus the shunt inductance for various
varactors at 2 GHz

C2 in series with an inductance L2 as shown in Fig. 2.5. The maximum inter-element

phase shift for such a structure is given by:

1 1 ωC max
ωC max − −
L1ω Lω t
Δθ = 2[arctan( ) − arctan( 1 )] (2.10)
G G

The graph of the maximum achievable phase shift versus shunt inductance L1 for

different varactors is shown in Fig. 2.6. For a given varactor capacitance, there is an

optimum shunt inductance value that maximizes the phase shift across the two nodes in

the circuit. This inductance value can be calculated based on the varactor maximum

capacitance Cmax and its tunability “t” as:

45
 Using the optimum capacitance value and no inductor [20]
Improved design with Cmax=10 pF
Improved design with Cmax= 5 pF
improved design with Cmax= 2 pF
Maximum Achievable Phase Shift (degree) 300

250

200

150

100

50

0
1 2 3 4 5 6 7 8 9 10
Varactor Tunability

Fig. 2.7. Maximum achievable inter-element phase shift versus the varactor tunability for
various varactor capacitances at 2GHz

2t
L opt = (2.11)
ω 2 C max ( t − 1 )

The maximum inter-element achievable phase shift is then given by:

⎡ ωC max ( t − 1 ) ⎤
Δθ max = 4 tan −1 ⎢ ⎥ (2.12)
⎣ 2Gt ⎦

The plots in Fig. 2.7 compare the maximum achievable inter-element phase shift

in the design of the extended resonance without any inductor [75] with the improved

46
  

Fig. 2.8. The impedance transformers can be employed to facilitate the implementation
of the extended resonance circuit

design for different varactor capacitance values. Low loss varactors based on MEMS,

BST and GaAs technologies with the typical tunabilities less than 2, 3 and 8,

respectively, can provide phase-shifts in the ranges of 10-20, 50-60 and 90-100 degrees

in the design based on [52], whereas the new design described here can achieve a

considerably larger phase shift employing the same varactors as seen in Fig. 2.7. This

improvement in the maximum achievable phase shift is even more significant if a

varactor with a larger capacitance value is used. Based on Eqn. 2.12, a varactor with a

larger tunability and capacitance value can result in a larger phase shift when an optimum

inductance is employed (Fig. 2.7).

In the extended resonance circuit shown in Fig. 2.5, varactors with different

capacitance values are required. In order to facilitate the implementation of the extended

resonance circuit, a different circuit topology (Fig. 2.8) is presented that can be designed

using similar varactors. In this circuit, the series combination of C2 and L2 (Fig. 2.5) is

replaced with a varactor Cp(V) in shunt with an inductor Lp inserted between two quarter-

47
wave transformers. The shunt combination of C1 and L1 (Fig. 2.5) is also replaced with a

quarter-wave transformer connected to a varactor CS(V) in series with an inductor LS, as

shown in Fig. 2.8. In order to achieve the maximum phase shift calculated in Eqn. 2.12,

the quarter-wave transformer is designed to transform the series combination of Cs and Ls

to the shunt combination of C1 and L1 with the optimum inductance value given by Eqn.

2.11. Furthermore, in order to allow the same varactor to be used through the entire

circuit, the impedances of the quarter-wave transformers are determined to provide the

susceptance ratios required in the extended resonance circuit. Moreover, the number of

bias voltages can be reduced dramatically in this structure. Two bias voltages can tune

the varactor capacitances through the entire circuit, one for tuning each of the varactors

CS(V) and one for tuning each Cp(V) (Fig 2.8).

The direction of the radiation beam of series-fed phased arrays changes with

frequency variation resulting in array beam squint. In order to investigate the frequency

response of the extended resonance phased array, several phased arrays are designed and

their beam squint characteristic versus frequency has been studied. The plots in Fig. 2.9

and Fig 2.10 show the inter-element phase shift and scan angle variation versus

frequency, respectively, for a phased array designed to have a scan angle of 15 degrees.

Three different circuits employing different varactors having tunabilities of 2, 5 and 10

are designed to achieve this scan angle. The results shown in the Fig. 2.9 and Fig. 2.10

suggest that beam squint is alleviated by increasing the varactor tunability. Based on Fig.

2.9 and Fig 2.10, as the tunability increases from 2:1 to 5:1, the beam squint slightly

improves however as the tunability increases beyond 5:1, there is minimal improvement

in frequency response of the phased array. The Fig. 2.11 and Fig. 2.12 show the inter-

48
 15
C=0.41 pF, t=2

Phase(degree)
10 C=0.26 pF, t=5
C=0.23 pF, t=10
5

Phase shift
0

-5

-10

-15
1.90
-5% 1.95
-2.5% 2.00
0 2.05
+2.5% 2.10
+5%
freq, GHz
Frequency variation

Fig. 2.9. Inter-element phase variation versus frequency for the 15 degree-scan range
phased array using different varactors

4
C=0.41 pF, t=2
C=0.26 pF, t=5
(degree)

2 C=0.23 pF, t=10

Angle

0
Scan Angle

-2

-4
1.90
-5% 1.95
-2.5% 2.00
0 2.05
+2.5% 2.10
+5%
freq, GHz
Frequency variation

Fig. 2.10. Scan angle variation versus frequency for the 15 degree-scan range phased
array using different varactors

49
 15
C=0.83 pF, t=2

Phase(degree)
10 C=0.52 pF, t=5
C=0.46 pF, t=10
5

Phase shift
0

-5

-10

-15
1.90
-5% 1.95
-2.5% 2.00
0 2.05
+2.5% 2.10
+5%
freq, GHz
Frequency variation

Fig. 2.11. Inter-element phase variation versus frequency for the 30 degree-scan range
phased array using different varactors

2
Angle

-2

-4

-6
1.90 1.95 2.00 2.05 2.10
freq, GHz

Fig. 2.12. Scan angle variation versus frequency for the 30 degree-scan range phased
array using different varactors

50
 30

20

10
Phase
0

-10

-20

-30
1.90 1.95 2.00 2.05 2.10
freq, GHz

Fig. 2.13. Inter-element phase variation versus frequency for the 90 degree-scan range
phased array using different varactors

10

5
Angle

-5

-10
1.90 1.95 2.00 2.05 2.10
freq, GHz

Fig. 2.14. Scan angle variation versus frequency for the 90 degree-scan range phased
array using different varactors

51
element phase shift and scan angle variation versus frequency for an array designed to

achieve a scan angle of 30 degrees. Compared to the graphs given the array having a scan

angle of 15 degrees, the beam squint degrades as the scan angle increases. Here, again the

beam squint is alleviated by increasing the varactor tunability. The plots of inter-element

phase shift and scan angle variation versus frequency for an array having a scan angle of

90 degrees are shown in Fig. 2.13 and Fig. 2.14.

The phased array frequency bandwidth requirement depends on the application

which the phased array is intended. Long range automotive radars for automotive cruise

control and crash warning systems require a bandwidth of approximately 1% at the center

frequency of 24 GHz. Furthermore, these radars are required to have a beam as narrow as

3 degrees which should be steered within scan range of 15 degrees around broadside.

Based on Fig. 2.12, for such a scan angle, the beam of the extended resonance phased

array tilts within a range as narrow as +/-0.4 degrees. Such a low frequency beam squint

capability demonstrates extended resonance phased array as a promising solution for

automotive long range radars [77].

2.4 Modular Extended Resonance Phased Array

One of the challenges in the area of phased array systems is how to extend very

large phased arrays. The maximum gain of an antenna array can be calculated as [46]:

G array = G antenna + 20 log 10 N (2.13)

where N is the number of array elements.

52
 Fig. 2.15. The modular design of transmit extended resonance phased array

Based on Eqn. 2.13, increasing number of array elements increases the gain of

antenna array. Moreover, the spatial selectivity of phased array can be enhanced by

increasing number of array elements. Consequently, the noise performance of phased

array system if used in receive mode can be enhanced. Furthermore, increasing the gain

of phased array alleviates the output power requirement of power amplifier if phased

array used in transmit mode [46]. In general, increasing the number of array elements

increase the design complexity of phased array. Power combing among a large number of

radiating elements can be exceedingly complex. Furthermore, the lossy passive signal

distribution networks can cause an excessive amplitude drop along the feed network of

phased arrays.

In the extended resonance circuit, the shunt susceptance values increase as odd

multiples of the first shunt susceptance value across the circuit, therefore as the number

of array elements increases, the value of the required shunt susceptances can be

prohibitively large. Furthermore, the loss of passive elements accumulates across the

circuit and degrades the uniform array excitation as the number of array elements is

increased. In order or address the aforementioned design issues, a modular approach to

53
the design of extended resonance phased array is presented. In this design, several

identical extended resonance circuits are placed in tandem through amplifier stages (Fig.

2.15), that is the last antenna at each module is replaced with an amplifier. After boosting

the signal level, it is injected into the next module. Therefore, the same design can be

used to extend the number of array elements realizing scalable phased array design. The

design complexity associated with large phased array design can be reduced by using the

scalable phased array presented here.

The amplifier stages are designed to inject signals with equal power levels into

each module in order to maintain a uniform power distribution across the entire antenna

array. Therefore, the gain of the amplifier can be determined by Eqn. 14 assuming a

lossless circuit.

G amplifier = 20 log 10 N (2.14)

Here, N is the number of array elements.

In reality, the amplifying stages should compensate for the circuit loss through the

circuit as well. The amplifier should be matched to G, the admittance of the antennas, in

order to maintain the uniform power distribution across each module. The output of the

amplifier is matched to NG to avoid any mismatch and reflection at the input of each

extended resonance module.

54
2.5 Heterodyne Modular Extended Resonance Phased Array

In order to apply extended resonance technique to microwave and millimeter

wave applications, low loss tunable elements required to operate at the mentioned

frequencies. Designing low loss tunable elements at such high frequencies can be

challenging. In this section, the design of an extended resonance phased array using a

heterodyne concept is presented. Exploiting heterodyne concept for phased array design

allows the tunable elements to operate at frequencies much lower than the actual

operating frequency of phased array.

2.5.1 Overview of Heterodyne Modular Extended Resonance Phased Array

As depicted in Fig. 2.16, the heterodyne phased array is composed of identical IF

power dividing/phase shifting and LO power dividing extended resonance modules. The

LO power dividing network, designed based on the extended resonance technique,

distributes the LO signal equally to the LO ports of the mixers. The IF-network is a

tunable extended resonance network that, in addition to uniformly distributing the IF-

signal to the mixers’ IF-ports, controls the phase of the IF signals. Thereby, the phased

array beam can be controlled by tuning the IF network. The identical IF and LO extended

resonance modules are connected in tandem through amplifying stages. The amplifying

stages are designed to inject the same power level into each module in order to maintain a

uniform power distribution across the entire networks. Furthermore, through heterodyne

mixing of the IF and LO signals, the insertion phase of the amplifying stage in the IF

network is canceled out by the insertion phase of the amplifying stage at the LO network.

55
 The extended resonance modules provide a linear phase progression across their

output ports [75]. Here, the initial phase progression across the IF ports is canceled out by

the phase progression across the LO ports to set the beam corresponding to the mid-point

of the scan range at broadside. In order to steer the beam, the phase shift between the RF

signals can be tuned by utilizing varactors in the IF or LO networks. Phase shifting at the

IF stage is preferred due to the better performance of varactors at lower frequencies. The

tunable IF modules are realized based on a modified extended resonance circuit topology

discussed previously.

The signal at an IF port can be presented by Eqn. 2.15.

S IF ( n ) = AIF cos( ω IF + nθ 0 + nΔθ ) (2.15)

Furthermore, the signal at the corresponding LO port is given by Eqn. 2.16.

S LO( n ) = ALO cos( ω LO + nφ 0 ) (2.16)

Therefore, the signal fed to the corresponding antenna element can be determined

according to Eqn. 2.17 assuming the mixer gain to be Amixer.

S RF ( n ) = Amixer AIF cos( ω IF + 2ω LO + 2 nφ 0 + nθ 0 + nΔθ ) (2.17)

Therefore, the frequency of RF transmit signal is given by Eqn. 2.18.

f RF = f IF + 2 f LO (2.18)

56
Fig. 2.16. The proposed phased array is a heterodyne-mixing structure that exploits the
extended resonance technique to achieve a uniform power distribution and to control the phase
progression across the phased array

57
 Furthermore, the initial phase progression across the IF ports is canceled out by

the phase progression across the LO ports allowing scanning around the broadside.

Therefore, the relationship given by Eqn. 2.19 should be met to set the initial beam

corresponding to the mid tuning range at broadside.

− θ0
φ0 = (2.19)
2

2.5.2 IF Circuit design

The IF-network, designed at 2 GHz, is composed of two tunable, four-port

extended resonance power dividing modules cascaded through an amplifying stage. The

IF power dividing module is shown in Fig. 2.17. In order to achieve an equal power

division through the entire IF-network, the designed amplifying stage has a gain of 7 dB,

matched to 50 ohm at the input and 10 ohm at the output. In addition to IF signal

distribution, the power dividing modules provide the required phase shift to steer the

beam. The phase shift can be controlled by tuning the varactors Cs(V) and Cp(V) as shown

in Fig. 2.6. The capacitance of the varactors Cs(V) and Cp(V) can be tuned from 0.2 pF to

1.2 pF and 0.4 pF to 2.4 pF, respectively, as the bias voltages change from 0.5 volts to 12

volts. The Cp(V) bias voltage VB1 is applied through a single bias circuit at the input of

each block, while separate, but identical, bias circuits are designed to tune the capacitance

of varactors Cs(V) using bias voltage VB2. The RF-choke inductors used in the bias circuit

design have an inductance of 154 nH, while the capacitance of the DC-blocks are 850 pF.

The inductors used in the design are LS=3.3 nH and LP= 9.5 nH. The characteristic

impedances of the quarter-wave transformers range from 30 ohms to 90 ohms through the

58
Fig. 2.17. The IF extended resonance module at 2 GHz

59
 Fig. 2.18. The inter-element phase shift can be controlled by tuning the varactors capacitances

entire design. In order to provide isolation between the power divider and mixers circuits,

the buffer amplifiers are employed at the IF stage. The graph of the resultant inter-

element phase shift versus the capacitance of varactor Cs(V) and Cp(V) is shown in Fig.

2.18.

2.5.3 LO Circuit design

The LO network is composed of two cascaded extended resonance power

dividing modules designed at 11 GHz. The LO power dividing module is shown in Fig.

2.19. The sub-harmonic mixers double the frequency of the LO signals and up-convert

the 2 GHz IF signals to 24 GHz transmit signals. In the design of the LO power dividing

modules, microstrip lines are utilized to achieve a compact, low loss structure at the

design frequency. Furthermore, a matching circuit is used to improve the impedance

60
Fig. 2.19. The LO extended resonance module at 11 GHz

61
 Fig. 2.20. The patch antenna designed for center frequency of 24 GHz

match at the mixers LO-ports. Moreover, the phase progression across the LO-network

cancels out the phase progression across the IF network to set the beam at broadside.

2.5.4 Antenna design

Microstrip patch antennas are used as radiating element in the presented phased

array. The patch antenna at the center frequency of 24 GHz was designed based on the

procedure given in [78]. Based on this procedure, the width of the antenna is given by:

c 2
W= (2.20)
2 f 0 ε reff + 1

62
 Fig. 2.21. The radiation pattern of the patch antenna at 24 GHz

where c is the speed of light, f0 is the resonant frequency of the microstrip patch antenna,

and εr is the dielectric constant of the substrate. The effective dielectric constant of the

microstrip antenna is calculated according to Eqn. 2.21.

−1 / 2
εr +1 εr −1 ⎡ h⎤
ε r − eff = + 1 + 12 ⎥ (2.21)
2 2 ⎢⎣ W⎦

where h is the substrate thickness. Once W and εr-eff is calculated, the extension length of

the microstrip patch due to the fringing effects can be calculated using.

63
 Fig. 2.22. Return loss of the antenna at 24 GHz

W
( ε r _ eff + 0.3 )(
+ 0.246 )
ΔL = h × 0.412 h (2.22)
W
( ε r _ eff − 0.258 )( + 0.8 )
h

Therefore, the length of the microstrip patch is given by:

c
L= − 2ΔL (2.23)
2 f 0 ε r − eff

Preliminary design of a 24 GHz microstrip patch antenna was done using the

given equations and the substrate parameters. Then, the antenna was simulated using

64
 Fig. 2.23. The photograph of the IF circuit

Ansoft HFSS and based on simulation results, the dimensions were optimized to achieve

the correct resonant frequency. The final patch antenna with its dimension is shown in

Fig. 2.20. The simulated radiation pattern of the final microstrip patch antenna is shown

in Fig. 2.21. The antenna is designed to be matched to 50 ohm. The simulated return loss

of antenna is shown in Fig. 2.22. The 10 dB return loss-bandwidth of the antenna is

approximately 500 MHz at the center frequency of 24 GHz.

2.6 Fabrication and Measurement results

A seven-element phased array is fabricated as shown in Fig. 2.23. As can be seen,

the IF-circuit is fabricated on a RO4003C from Rogers Corporation with a thickness of

20 mils and a dielectric constant of 3.55. The flip-chip GaAs varactor MA46H120 from

MACOM is used for phase shifting. The varactors have an approximate quality factor of

65
Fig. 2.24. The measured array factors of the IF-network compared to the simulation

66
 Fig. 2.25. The photograph of the LO circuit and the patch-antenna elements together with
their feed lines

50 at a bias voltage of 4 volts and 5:1 tunability at 2 GHz. The amplifier ERA1+ from

Mini-circuits is employed to cascade the IF power dividing modules. The same amplifier

is used as the buffer at the IF ports. The IF-circuit can be used as a stand-alone circuit for

beam steering at 2 GHz. The S-parameters of the IF ports are measured at various bias

points and the calculated array factors are shown in Fig. 2.24. A scan angle of

approximately 60 degrees can be achieved at the IF-circuit while the side lobes are

maintained below -10 dB. The 3-dB beam width of the array factor varies from 15

degrees to 18 degrees within the entire scan range.

The LO-circuit together with the antenna elements are fabricated on a separate

circuit board as shown in Fig. 2.25. The substrate used is Rogers RO4350 with a
67
  

Fig. 2.26. The phased array radiation pattern measurement setup

thickness of 10 mils and a dielectric constant of 3.66. The LO signal at 11 GHz provided

by a signal generator is applied to the input of the LO circuit. The extended resonance

modules in the LO circuit are cascaded through RFMD LN300 amplifiers. The Sub-

harmonic mixers HMC264LM3 with integrated LO frequency doublers from Hittite

Microwave are used in the phased array. The LO amplifiers are included in these mixers

obviating the need for a high LO power.

The IF signals from the IF circuit board are injected into the IF ports of the mixers

on the LO circuit board. The outputs of the mixers are connected to patch antennas at 24

68
Fig. 2.27. The measured array patterns at 24 GHz compared to the array factors calculated
using the simulated S-parameters

69
 Fig. 2.28. Measured input reflection coefficient of the IF circuit versus frequency at different
scan angles

GHz. The patch-antennas are placed λ/2 apart and each has a gain of 5 dBi. The

dimensions of these antennas are 4 mm by 3.28 mm.

The measurement set-up in an anechoic chamber is shown in Fig. 2.26. A Ka

band horn antenna MFR-01456 is used as the receive antenna to measure the array

pattern. The receive-signal is amplified by an LNA before being measured by a spectrum

analyzer and then the computer processes the data to determine the radiation pattern. The

computer also controls the rotation angle of the phased array. The radiation patterns

corresponding to the broadside and maximum scan angles are measured. The measured

results are compared with the array factors calculated using the simulated S-parameters as

shown in Fig. 2.27. A scan angle of approximately 50 degrees is achieved. One source of

70
 Fig. 2.29. The array factor based on measured S-parameters at 2 GHz versus frequency

discrepancy between the measured array patterns and the calculated array factors is the

radiation from the meandered microstrip lines.

Any modulation in a heterodyne system usually happens at the IF stage.

Therefore, it is necessary to understand the effect of frequency variation within the IF-

path on the performance of the proposed phased array system. The input return loss and

the array factor are measured at the IF-circuit at different frequencies. As can be seen in

Fig. 2.28, the return loss of the IF-circuit is maintained greater than 10 dB at a bandwidth

of above 20% in the entire scan range. Based on the measured results, the beam tilts

approximately within +/- 7 degrees within a bandwidth of 100 MHz. The change in peak

gain of the array is approximately 1 dB while the side lobes are maintained below 10 dB
71
in the 100 MHz bandwidth (Fig. 2.29). Symmetrical design of the series fed phased array

can alleviate the beam squint.

TABLE 2.1
Comparison between the presented extended resonances phased array and the published series-
fed phased arrays
[79] [80, 81] [82] [52] [83] Presented
Phased
Array
Number of 4 5 patches 30 leaky 4 patches 4 patches 7 patches
antenna patches wave
elements
Center 2.45 5.8 GHz 3.33 GHz 2 GHz 2.4 GHz 2 GHz
frequency GHz
Scan Range 30 22 60 20 49 60
degrees degrees degrees degrees degrees degrees
Number of 6 1 1 4 3 2
tuning voltages
Return loss 1.02% 4.6% NA NA 3% 20%
bandwidth
Beam squint NA 5.6° per 2.7° per NA 0.4° per 2.8° per
1% 1% 1% 1%
frequency frequency frequency frequency
variation variation variation variation
Gain variation NA 0.4 dB 13 dB 1.8 dB 1.5 dB 1.8 dB
within scan
range

Table 2.1 summarizes the performance of the proposed electronically steerable

antenna array against other series-fed steerable arrays presented in the literature. The

proposed antenna array achieves a much wider electronic scan-angle range compared to

the other designs shown in the table 2.1. The design proposed in [85] can achieve a

comparable scan range at the price of 13 dB array gain variation, whereas, the extended

resonance phased array demonstrates a gain variation of only 1.8 dB within the entire

scan range. Furthermore, the extended resonance phased array requires a fewer number of

72
tuning voltages to control the beam of phased array compared to the works presented in

[82], [52], [86]. The return loss bandwidth of the extended resonance phased array is

much broader than any of the other designs while it demonstrates relatively low beam

squint capability as well. The beam squint can be alleviated in the extended resonance

phased array by reducing its scan range.

2.7 Conclusion

A new design of phased array based on extended resonance technique is

presented. The extended resonance technique is capable of reducing the complexity of

phased arrays by achieving phase shifting and power dividing within a single network. A

new circuit topology to enhance the scan range of extended resonance is reported and the

maximum achievable inter-element phase shift across the tunable extended resonance

network is studied. Moreover a modular approach to realize a scalable phased array

design is proposed which can significantly reduce design complexity associated with

large phased arrays. Furthermore, a 24 GHz modular transmit phased array based on the

extended resonance technique is reported. This technique is used along with heterodyne-

mixing concept which allows the design to benefit from the better performance of tunable

elements like varactors at lower frequencies. Finally a seven-element 24 GHz modular

transmit phased array is fabricated and tested. The measured results show a good

agreement with the simulation.

73
 Chapter 3

A New Bi-Directional Series-Fed Phased Array

3.1 Introduction

As mentioned before, the large number of required phase shifters and their high

circuit complexity is one of the main reasons behind the high cost and complexity of

phased arrays. Furthermore, the cost and complexity of the individual phase shifters in a

phased array depends on the required phase shift. As an example, one type of phase

shifters so called “varactor-loaded-transmission line” is shown in Fig. 3.1. In this type of

phase shifter, the maximum achievable phase shift is determined by the number of

varactor and inductor stages. Therefore, overall phase shifter complexity is directly

proportional to the maximum phase shift required from phase shifter. By designing

phased arrays that demand less phase shift from phase shifters, the cost and complexity of

phased array can be reduced.

74
 Fig 3.1. The complexity of varactor loaded transmission line is determined by the amount
of phase shift required from the phase shifter

The required amount of phase shift from a phase shifter can depend on the design

of the array feed network. In this chapter, different types of array architectures are

discussed and their required amounts of phase shift are compared. A new bi-directional

phased array is introduced which demands less phase shift from phase shifters compared

to any of the conventional phased array designs. The detailed design of the feed network

is given in this chapter. Furthermore, a new compact phase shifter design is presented in

this chapter as well. Utilizing this phase shifter within the proposed bi-directional phased

array allows the array beam to be steered by a single control voltage. The entire phased

array is fabricated and its measurement result is reported.

3.2 Overview of phase Shift Requirement in Different Types of Phased

Arrays

In general phased arrays are designed based on either a parallel or a serial feed

network [84]. In parallel feed networks, which are often called corporate feeds, the input

75
Fig 3.2. Phased arrays using parallel feed network require maximum phase shift of (N-1)θ
from phase shifters to achieve inter-element phase shift of θ

Input

Fig 3.3. Series-fed phased array with phase shifters inserted in parallel require maximum
phase shift of (N-1)θ from phase shifters to achieve inter-element phase shift of θ

Fig 3.4. Series-fed phased array with phase shifters inserted in series require maximum
phase shift of θ from phase shifters to achieve inter-element phase shift of θ

76
 Fig 3.5. Bi-directional series-fed phased arrays require maximum phase shift of θ/2 from
phase shifters to achieve inter-element phase shift of θ

signal is divided in a corporate tree network to all the antenna elements as shown in Fig.

3.2. In phased arrays using corporate feed network, the phase shift at each individual

element is controlled separately through a phase shifter. In general, for an N-element

phased array with a corporate feed network, to achieve inter-element of θ, maximum

phase shift of (N-1)θ is required from phase shifters. Series feed networks are simpler,

more compact and with lower feed line losses compared to parallel networks [85]. In a

series-fed array the input signal, fed from one end of the feed network, is coupled serially

to the antenna elements. Also the other end of the network is usually terminated with a

matched load. In serially fed arrays, the phase shifters can be inserted either in parallel

(Fig. 3.3) or in series along the feed line as depicted in Fig. 3.4, which demands a reduced

phase shift for a given scan angle. In an N-element serially fed array, placing the phase

77
shifters in series rather than in parallel reduces the required amount of phase shift by a

factor of (N-1) [34]. To further reduce the required amount of phase shift, such arrays can

be bi-directionally fed, as shown in Fig. 3.5. Bi-directionally series-fed arrays have been

implemented using both hybrid [79, 86] and integrated circuits [87]. The bidirectional

series-fed phased array presented in [79] uses phase shifters having variable characteristic

impedances that are adjusted according to the feeding direction. In [86], two different

beams are formed by switching the source between the two opposite ends of a dual feed

travelling patch antenna array. The authors in [87] combined a bi-directional series feed

network with a parallel-fed architecture in order to reduce the required phase tuning

range. Finally, the bi-directional fed series fed array presented in this chapter reduces the

phase shift required for a given scan angle to half of the conventional series fed arrays

illustrated in Fig 3.4. The detailed design of this phased array is given in the next section.

3.3 A New Bi-Directional Series-Fed Array to Reduce the Required

Amount of Phase Shift

3.3.1 Bi-Directional Phased Array Overview

In this chapter, we propose a new bi-directional series-fed phased array approach

that can reduce the required phase shift to half of the phase shift needed in the

conventional serially fed arrays shown in Fig. 3.4. Furthermore, unlike the approach

described in [79], the phased array presented here does not require any adjustment in the

feed network as the feeding direction of the array is changed. The approach reported here

allows the same phase shifter design to be used throughout the entire phased array

substantially reducing array design complexity. Since a compact, low complexity phase

78
Fig 3.6. The new bidirectional phased array utilizes directional couplers, switches and phase
shifters with a single control voltage to achieve beam steering.

79
shifter can be used throughout the array, array beam can be controlled by a single bias

voltage, obviating the need for complex bias circuits. In the next section, the theory of

bidirectional phased array is presented. In the following sections, the implementation and

measurement results of an eight-element phased array designed based on the presented

architecture are discussed.

3.3.2 Operation Theory of the Bi-directional Series-Fed Phased Array

A new bi-directional array composed of identical couplers and switches connected

in tandem along with phase shifters is shown in Fig. 3.6. All of the phase shifters are

identical and simultaneously tuned by a single control voltage through the feed network.

By using a matched single pole, double throw (SPDT) switch, the array can be fed from

either ends. If the phase shifters provide an inter-element phase shift ranging from 0 to θ,

the array radiation beam can be steered from 0 to φ1=arcsin(θ/π) assuming that the array

elements are λ/2 apart. When the feed direction is switched, the phase shifters provide an

inter-element phase shift ranging from –θ to 0 allowing for the beam to be steered from 0

to φ2=-arcsin(θ/π). As a result, array beam can be steered by 2arcsin(θ/π).

3.3.3 Bi-Directional Feed Network Design

As the array feed direction is switched, it is important to maintain the array

directivity, beamwidth and side-lobe levels the same. In order to achieve this, all of the

couplers used in the array are identical. Furthermore, using the identical couplers obviates

the need to design each individual array element separately, thereby reducing design

complexity. In order to determine the optimum coupling factor, its effect on the array

80
Fig 3.7. Array directivity versus the coupling factor for different array sizes

Fig 3.8. Increase in array beamwidth compared to a uniformly excited array versus the
coupling factor for different array sizes

81
amplitude excitation is investigated. Assuming the phase shifters and the couplers are

ideal and lossless and all the couplers have the same coupling factor “C”, the normalized

signal level at each antenna element is given in Eqn. 3.1.

_
S n = (1 − C ) n −1 C (3.1)

where “n” is the antenna element number.

According to Eqn. 3.1, the signal amplitude at each array element drops along the

feed-line resulting in a linear amplitude tapering that affects the array directivity and

beam-width. To investigate the effect of amplitude tapering on the array factor, the array

directivity “D” has been calculated (2). As shown by (2), array directivity depends on

the number of array elements “N” as well as the coupling factor “C”. The antenna

elements here are assumed to be omnidirectional with λ/2 spacing.

1 ⎡ N ⎤
C2 ⎢ 1 − (1 − C ) 2 ⎥
D= 1 ⎢ N ⎥ (3.2)
⎢
1 − (1 − C ) ⎣ 1 + (1 − C ) 2
2
⎥⎦

Based on Eqn. 3.2, directivity of a bidirectional array having 5, 10 and 15

elements versus the coupling factor “C” is plotted in Fig. 3.7. As can be seen, by

reducing the coupling factor, one can improve the array directivity. This also mitigates

the array beam broadening caused by amplitude tapering across the array (Fig. 3.8). At

the same time, the power dissipated in the matched termination within SPDT switch

increases as “C” is reduced. The relationship between the power dissipation normalized

82
 Fig 3.9. The percentage of input power dissipated within the matched SPDT switch versus the
coupling factor for different array sizes

to the input “Pdis”, the coupling factor “C” and the number of array elements “N” is

shown in Eqn. 3.3.

__
Pdis = (1 − C ) N (3.3)

A plot of the dissipated power percentage as a function of the coupling factor “C”

for various array sizes is given in Fig. 3.9. As can be seen, reducing coupling factor

increases the power dissipated at the matched termination within the SPDT switch at the

input to the array.

When determining the size of bidirectional array to meet a certain directivity goal,

the dissipated power “Pdis” should be also taken into account. In general, one can use

83
Fig 3.10. Directivity of an eight-element series-fed array compared to an eight-element
uniformly excited array at (a) scan angle of zero and (b) scan angle of thirty degrees.

84
Fig 3.11. Amplifier can be used in the receive phased array to achieve a particular
amplitude tapering

GN G1 GN-1 G2 G2 GN-1 G1 GN

Lp

C: Coupling factor

1-C

Fig 3.12. Amplifier can be used in the transmit phased array to achieve a particular
amplitude tapering

85
Fig. 6 to determine the numbers of array elements and a range of coupling factors to

achieve a required directivity. Subsequently, based on Fig. 3.9, the minimum number of

array elements can be chosen to maintain the dissipated power below a target goal. It can

be shown that by appropriately choosing the coupling factor; one can achieve a

directivity similar to that of a uniformly excited array using the same number of array

elements while less than 15% of the input power is dissipated within the matched SPDT

switch.

For the eight-element phased array presented in this paper, the coupling factor

is set to -7 dB. In this case 15% of the input power is dissipated in the SPDT termination.

The array factors for this design compared to a uniformly excited eight-element array for

scan angles of zero and thirty degrees are shown in Fig. 3.10. As can be seen, the

difference between the patterns is negligible. Therefore, by selecting an appropriate

coupling factor, one can achieve a radiation pattern which is very close to the theoretical

pattern for a uniformly excited array with the same number of array elements.

It should be noted that the emphasis of this chapter is only on the power

distribution and phase shifting. The phased array can contain low noise amplifiers and

power amplifiers as depicted in Fig. 5. Therefore, the performance of LNAs and PAs

ultimately should be considered to determine the dynamic range and efficiency of the

overall array.

Different amplitude tapering may be required across the array to meet a particular

sidelobe level requirement. Amplifiers can be used in the array for receive mode (Fig.

3.11) and transmit mode (Fig. 3.12) to achieve any amplitude tapering desired. Using

86
amplifiers also allows for compensating the effect of phase shifter insertion loss on the

array factor. In general, if array amplitude tapering of a1…an is desired across the array,

gain for array element n, Gn, can be calculated according to Eqn. 3.4 assuming the gain of

the first element to be G1 (shown in Fig. 3.11 and Fig. 3.12).

Gn L p n−1 an
=[ ] ( ) (4)
G1 1 − C a1

where Lp is the insertion loss through the phase shifter.

3.4 A New Phase Shifter Design

One of the key and essential components of most of phased arrays is phase shifter.

As mentioned before, a major complexity of phased array is attributed to the phase

shifters required. Furthermore, performance of phase shifters used in the phased array

significantly affects the performance of the entire phased array. Critical parameters in

phase shifter design are phase tuning range, RF insertion loss, variation of insertion loss

versus phase shift, power handling and bandwidth [88]. In the next section, a brief

overview of common approaches to design phase shifter will be given. In the following

section, design of a new compact phase shifter for the use in bi-directional phased array

will be discussed.

87
 Fig 3.13. A photograph and general diagram of loaded transmission line phase shifter

3.4.1 Overview of Common Phase Shifter Design Approaches

A variety of phase shifters have been investigated for use in phased arrays. The

most comment approaches to design phase shifters are based on a varactor loaded line,

reflective type, vector modulated or switch based phase shifters. In varactor loaded

transmission line design, the electrical length of the line is varied by tuning the varactors

(Fig. 3.13). Various varactor technologies have been used in this type of phase shifters

including MEMS varactors [89-91], BST varactors [92-94] or solid state varactors [95-

97]. Reflective type of phase shifter is shown in Fig. 3.14. In this type of phase shifter, a

88
Fig 3.14. A photograph and general diagram of reflective type phase shifter

Fig 3.15. A photograph and general diagram of vector modulated phase shifter

89
 Fig 3.16. A photograph and general diagram of switch line phase shifter

circulator or a couple is terminated with tunable reactances (Fig. 3.14). Tunable

reactances reflect the incident signal with a phase shift which depends on the value of the

reactance. The amount of phase shift can be tuned by using varactors to change the value

of reactance. Vector modulated phase shifter is another type of phase shifter that is

mainly used for integrated circuits (Fig. 3.15). In this phase shifter, signal is divided into

two paths with different gains and 90 degree phase difference. The signal at the output is

the vector sum of the two signals. It can be shown that the phase of the output signal

depends on the amplitude ratio of the two signals. Therefore, by tuning the amplitude

90
 Fig 3.17. The circuit diagram of the phase shifter.

ratio of the two signals, the phase of overall signal can be tuned. Switch based phase

shifters use switches to select different phase delays (Fig. 3.16). The phase delays can be

implemented based either on distributed or lumped elements.

3.4.2 Phase Shifter Design

The phase shifter used in the array is designed based on the extended resonance

technique as shown in Fig. 3.17 [98]. The phase shifter is assumed to be matched to

Z0=1/Y0. A varactor with a susceptance jB is connected in shunt at the input node. The

admittance seen at this node, Y0+jB, is then transformed to its conjugate value, Y0-jB, by

using a series reactance jX. The inductive value can be calculated based on Eqn. 3.5.

Finally, a similar varactor jB is added in shunt to cancel out the imaginary part of the

impedance.

91
 Fig 3.18. The required series inductance versus the varactor capacitance to maintain the same
voltage amplitude at the input and the output node

2B
X= 2
(5)
Y0 + B 2

Analysis of the circuit shows that for a fixed value of “B” the voltage at the input

node, Vin, and the voltage at the output node, Vout, have the same magnitudes but are

different in phase. The phase difference between Vout and Vin is given by Eqn 3.6.

B
Δθ = 2 arctan( ) (6)
Y0

92
 Fig 3.19. Maximum insertion loss versus maximum achievable phase shift when a fixed
reactance value is used.

In order to vary the phase shift, the susceptance jB is tuned by controlling the

varactor bias voltage. In general, as the varactors are tuned, the series reactance jX should

also be tuned to maintain the voltage magnitude at the nodes Vout and Vin the same.

However, for simplicity’s sake, the jX can be fixed at a value that allows the variation of

the amplitude ratio of the output voltage, Vout, to the input voltage, Vin to be minimized.

For a 2.4 GHz phase shifter with the characteristics impedance of 50 ohm, the required

series inductance versus the capacitance of varactor is plotted in Fig. 3.18. As can be

seen, at the point where the slope of the graph is zero, the required variation of

inductance versus varactor capacitance is minimal. Therefore, by choosing this optimum

value of series inductance, the variation of voltage amplitude ratio for

93
 -40

-60
phase(S(4,3))
phase(S(5,6))

-80

-100

-120

-140
0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
CX

Fig 3.20. Simulated phase shift through the phase shifter.

nodes Vout and Vin is minimal. In general, the optimum value of series reactance jXopt can

be calculated as Eqn. 3.7.

1
X opt = ωLopt = (7)
Y0

Phase shifters with different varactors are designed using the corresponding

optimum reactance given by Eqn. 3.7. The maximum phase shifter insertion loss within

the tuning range of varactor capacitance versus the maximum achievable phase shift is

shown in Fig. 3.19. As expected, as the maximum phase shift is

94
 0.0

-0.2

-0.4
dB(S(4,3))
dB(S(5,6)) -0.6

-0.8

-1.0

-1.2
0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
CX

Fig 3.21. Simulated insertion loss through the phase shifter

-10

-15
dB(S(3,3))
dB(S(6,6))

-20

-25

-30

-35
0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
CX

Fig 3.22. Simulated return loss of the phase shifter.

95
 Fig 3.23. A single voltage applied to the bias network at the input controls the phase shift
across the entire phased array

increased, the maximum insertion loss due to fixing the value of series reactance

increases as well.

Based on the design procedure described, a phase shifter is implemented using

varactor diodes and a lumped inductor at 2.4 GHz. The varactor capacitance used in the

design is tuned from 0.6 to 2.4 pF with the inductor value of 2.1 nH. The phase shift

through the phase shifter has been simulated both with ideal element as well as lumped

96
 Fig 3.24. A Photograph of the eight-element bidirectional phased array operating at 2.4 GHz

elements with Q = 50 (Fig. 3.20). The phase shift through the phase shifter can then be

tuned approximately from -45 to -150 degrees. The loss through the phase shift has also

been simulated (Fig. 3.21). The loss through the phase shifter due to the mismatch

doesn’t exceed 0.2 dB over the entire range. When varactor diodes and inductors with a

finite quality factor (Q = 50) are used, the maximum insertion loss increases to 1.2 dB. In

this case, the return loss is maintained below -15 dB in the entire tuning range. (Fig.

3.21).

In order to utilize this phase shifter in the design of the bidirectional phased array,

a fixed delay is introduced at each phase shifter by inserting two transmission lines as

97
 Fig 3.25. Measured phase shift through the phase shifter versus frequency for different
control voltages.

shown in Fig. 3.23. The phase shift across the entire array is controlled through the bias

circuit at the input to the bidirectional array (Fig. 3.23).

3.5 Fabrication and Measurement Results

An eight-element phased array at 2.4 GHz is designed and fabricated on Rogers

RO4003C with a dielectric constant of 3.55 and a thickness of 8 mils (Fig. 3.24). The

surface-mount couplers CP0805A2442BW from AVX with a coupling factor of 10 dB

are used in the design. The surface-mount power combiners from Mini-circuits (SP-

2U2+) combine the power from the output ports of the couplers.

98
Fig 3.26. Measured insertion loss through the phase shifter versus frequency for different
control voltages

Fig 3.27. Measured return loss of the phase shifter versus frequency for different control
voltages

99
Fig 3.28. Comparison between the measured and simulated array factors for the eight-element
phased array

100
 Fig 3.29. Input return-loss for different bias voltages

The phase shifter used in the design is fabricated and measured separately. The

varactor diode used in the phase shifter is MA46H120 from MA/COM with a tunability

of 4:1 and quality factor of 50 at the bias voltage of 12 volts and the design frequency.

Two varactor diodes are placed in shunt to provide a capacitance tuning range of 0.6 pF

to 2.4 pF as the bias voltage is varied from -0.1 volts to -16 volts. The lumped inductor

used in the design is 0403HQ-2N1XJL from Coilcraft with an inductance value of 2.1 nH

and a quality factor of 80 at the design frequency. A single bias voltage from 0.1 to 16

volts is used to tune the capacitance of the varactors. The measured phase shift versus

frequency is shown in Fig. 3.25. As can be seen, the measured phase shift through the

phase shifter can be tuned from -40 to -130 degrees, while its insertion loss varies from

0.2 dB to 1.8 dB over the tuning range (Fig. 3.26). The phase shifter return loss also

remains below 15 dB within the entire tuning range (Fig. 3.27).

101
 The array beam is controlled by a single bias voltage applied at the input to the

phased array. The S-parameters for the entire are measured as a function of varactors bias

voltage with the input signal switched between the two directions. The corresponding

array factors are then calculated assuming a linear array with λ/2 antenna spacing. A scan

angle of 60 degrees has been achieved as the tuning voltage is varied from -0.1 to -16

volts. The calculated array factors for scan angle of +/- 15 degrees and +/-30 degrees are

compared with the simulated results in Fig. 3.28. The 3-dB beamwidth is 17 degrees at

scan angle of 15 degrees and 14 degrees at the scan angle of 30 degrees while the side-

lobes are maintained at approximately below -10 dB. Based on the measurement result,

the input reflection coefficient S11 at the design frequency is better than -10 dB over the

entire scan range. As shown in Fig. 3.29, the reflection coefficient is maintained below -

10 dB within a bandwidth of approximately 400 MHz corresponding to a fractional

bandwidth of 17%.

Table 3.1 summarizes the performance of the proposed phased array against other

series-fed steerable arrays presented in the literature. The proposed antenna array

achieves a much wider electronic scan-angle range compared to the designs shown in

Table 3.1. The design proposed in [85] can achieve a comparable scan range at the price

of 13 dB array gain variation, whereas, the bi-directional series-fed phased array

demonstrates a gain variation of 2.7 dB in the entire scan range. Furthermore, the bi-

directional series-fed phased array requires only one tuning voltage to control the phased

array beam. The return loss bandwidth of the bi-directional phased array described here is

much broader than any of the other designs.

102
 TABLE 3.1
Comparison between the bi-directional series-fed phased array presented in this chapter and the
published series-fed phased arrays
[79] [80, 81] [82] [52] [83] The
Presented
Phased
Array
Number of 4 5 30 leaky 4 4 8 elements
antenna elements elements wave elements elements
elements
Center 2.45 5.8 GHz 3.33 2 GHz 2.4 GHz 2.4 GHz
frequency GHz GHz
Scan Range 30 22 60 20 49 60 degrees
degrees degrees degrees degrees degrees
Number of 6 1 1 4 3 1
tuning voltages
Return loss 1.02% 4.6% NA NA 3% 17%
bandwidth
Gain variation NA 0.4 dB 13 dB 1.8 dB 1.5 dB 2.7 dB
within scan
range

3.6 Conclusion

A new series fed bi-directional phased array is reported. The phased array uses a

bi-directional series feed network to reduce the required phase tuning range. As the result,

complexity of the phase shifters required in the array can be reduced resulting in overall

phased array cost and complexity reduction. A review of different topologies of phased

array and the required phase shift is given in this chapter. The operation theory of the bi-

directional phased array is discussed and the required phase shift for a given scan range is

determined. A general design procedure for a bidirectional phased array feed network is

presented and effect of design parameters on the array factor is discussed. Moreover,

103
review of common approaches to design phase shifters are given in this chapter and a

new, compact phase shifter is designed and utilized in the phased array. The phase shifter

has been fabricated and measured. The overall phased array architecture allows one to

steer the radiation beam by a single control voltage obviating the need for complex bias

circuits. An eight-element phased array based on the presented design is fabricated and

tested. The agreement of the measured results with the simulation verifies the design.

104
 Chapter 4

A Novel Design of Low Complexity Phased Arrays

4.1 Introduction

In this chapter, we propose a new approach to the design phased arrays with low

complexity. Unlike the conventional design of phased arrays requiring a separate phase

shifter for each antenna element (Fig. 4.1), in the proposed design, the phase progression

across the antenna elements can be controlled by using only a single phase shifter (Fig.

4.2). As a result, the cost and complexity of the phased arrays can be significantly

reduced based on this design. The proposed phased array is composed of couplers,

amplifying stages, power dividers and two phase shifters. The phase shifters are placed at

the two opposite ends of a serially fed array. In the next sections the proposed design of

the phased array is discussed. Finally in the last section, the implementation and

measurement results for an eight-element phased array operating at 2 GHz are presented.

105
Fig. 4.1. Conventional design of phased arrays requires a separate phase shifter for each
antenna

Fig. 4.2. The proposed phased array design requires a single phase shifter placed at the
opposite end of a serially excited antenna array to the input

106
4.2 The Concept for a phased array with only a single phase shifter

In most of the existing phased array designs, the phase of the signal at each

antenna element is separately controlled by a phase shifter. However, in the presented

approach, the phase of signals in the entire phased array can be controlled by a single

phase shifter. The block diagram shown in Fig. 4.3 illustrates the concept of the proposed

approach. The signal at each radiating-element is the vector sum of two signal

components, ai and bi provided by distributing the input signal through two feed networks

with unequal power division. The signals across each feed network are in phase while the

phase difference between the two signal paths is controlled by a single phase shifter. The

signal phase shift at each radiating element depends on the phase and amplitude ratio of

the two signal components ai and bi at each radiating element. The phase shift θi at

antenna element Ei is given by:

⎛ ⎞
⎜ ⎟
⎜ Sinφ ⎟
θ i = arctan (4.1)
⎜ ai ⎟
⎜ b + Cosφ ⎟
⎝ i ⎠

where Ф is the signal phase shift through the phase shifter as shown in Fig. 4.3

107
 a1 > a2 > …> an b1 < b2 < …< bn

Fig. 4.3. The block diagram of phased arrays based on the proposed approach requiring a
single phase shifter throughout the entire phased array

Fig. 4.4. Phase progression in the entire array can be controlled by tuning the phase of one
component of the signals at the antenna elements

Based on Eqn. 4.1, the phase shift at each antenna element Фi is a function of

ai/bi. Therefore, by providing different amplitude ratios for ai and bi, a different phase

108
 Fig. 4.5. The block diagram of phased arrays based on the proposed approach requiring a
single phase shifter throughout the entire phased array

shift can be achieved across the array with a single phase shifter. The amplitude ratio for

the two signals at each array element can be determined to provide an appropriate phase

progression across the array. In this design, the first signal component fed into each

antenna element progressively decreases, while the second signal component

progressively increases across the array. The vector diagram shown in Fig. 4.4 illustrates

how the phase progression is achieved across the phased array.

The maximum achievable scan angle θmax in phased arrays designed based on this

approach can be approximated using Eqn. 4.2. The equation is derived assuming the

phase shifter can provide a maximum phase shift of 360 degrees. Therefore, the

maximum inter-element phase shift can be determined by splitting the maximum

achievable phase shift from phase shifter by the number of array elements.

109
 180 2
θ max = a sin( ) (4.2)
π N

N is the number of antenna elements.

As can be seen, the maximum achievable scan range is dependent on the number

of antenna elements used in the array. As the number of antenna elements increases, the

maximum scan range decreases. For applications where a wide scan range is not needed,

the presented method can be exploited to substantially reduce the number of phase

shifters. For example, long range automotive radars for automatic cruise control or pre-

crash warning systems require radiation beams as narrow as 3 degrees which should be

steered within 15 degrees around the broadside. In order to achieve such a beamwidth, at

least 24 antenna elements are needed to be used in the design of phased array. Based on

Eqn. 4.2, to achieve 15 degrees of scan angle, one phase shifter is required for each 9

antenna elements. Assuming the modular approach introduced in the chapter 2 can be

used, it means 3 phase shifters are required for the entire phased array designed based on

the presented approach. A low number of phase shifters required compared to the

conventional phased array designs which usually require a single phase shifter per each

antenna element renders this approach as a promising solution for the design of long

range automotive radars.

The single phase shifter phased array design methodology presented here is

general and can be exploited to construct both active as well as passive phased arrays.

Furthermore, both receive mode and transmit mode phased arrays can be realized based

on the presented approach.

110
Fig. 4.6. The first signal components are achieved as the signal is traveling away from input

Fig. 4.7. The second signal components are achieved as the phase shifted signal is reflected
back into the feed network

111
Fig. 4.8. The overall signal the each antenna element is achieved by combining two signal
components

Fig. 4.9. Approximately linear phase progressing can be achieved across the array

112
One circuit implementation of the presented approach is discussed in the next section.

4.3 Phased array architecture

The circuit diagram of the proposed phased array is shown in Fig. 4.5. The phased

array is composed of a serial feed network and a phase shifter. The opposite end of

serially fed array is connected to a phase shifter. The phase shifter, which consists of

varactor diodes and inductors, reflects the incident signal back after providing a phase

shift. In the next sections, the serial feed network and phase shifter designs are presented.

4.3.1 Feed network design

As can be seen in Fig. 4.5, the series feed network is composed of directional

couplers and amplifiers. The signal at each antenna port is the vector sum of the two

signal components travelling in opposite directions. First signal components are

generated when the input signal is travelling away from the input. The second signal

components are generated when the input signal is phase shifted and reflected back into

the feed network. The first signal component is coupled through the directional coupler’s

coupled port, while the second signal component is coupled through the isolated port of

the coupler. The first signal component at each antenna element has a fixed phase which

is maintained for all the antenna elements. However the second signal component has a

variable phase controlled by the phase-shifter. By tuning the phase of the second signal

component, a variable progressive phase-shift across the sub-array can be achieved. The

amount of the phase shift at each antenna element depends on the relative amplitude of

the two signal components. In an array with linear phase progression, the closer an

113
 Fig. 4.10. The amplitude variation at the antenna elements as the phase progression is tuned

antenna element is to the phased array input, the larger the amount of phase shift it would

require. Therefore, as the amplitude of the first component of the signal decreases, the

amplitude of the second component increases. To provide proper signal amplitudes across

the array, 3 dB hybrid couplers are used in the design. Furthermore, in order to

compensate for the amplitude difference between the input signal and the reflected signal

by the phase shifter, amplifiers A1…A4 with appropriate gains are incorporated in the

design. The phase shift at each antenna is shown in Fig. 4.9. As can be seen by choosing

appropriate coupling factor, approximately linear phase progression can be achieved

across the array.

114
 Fig. 4.11. Circuit diagram of the phase shifter

The signal at each antenna port is the vector sum of the two signal components,

therefore by tuning the phase of one component the overall amplitude of vector sum

varies. The amplitude variation at the antenna elements are shown in Fig. 4.10. A

maximum amplitude variation of 7 dB is required to keep the uniform array excitation.

To maintain the amplitude of the signal at each array element, variable-gain amplifiers

B1…B4 are employed before the antenna elements.

4.3.2 Phase shifter design

The circuit structure of the reflective phase shifter is shown in Fig. 4.11. The

phase shifter circuit is designed using several varactor diodes and lumped inductors. The

phase shifter behaves like an open circuited transmission line with an adjustable electrical

115
Fig. 4.12. Phase shift of the signal reflected by the phase shifter

Fig. 4.13. Loss of the signal reflected by the phase shifter

116
Fig. 4.14. Two sub-array are connected to form a symmetrical structure fed from the center

117
length. The effective electrical length of the line can be controlled by tuning the varactor

capacitance. The amount of phase shift can be controlled by varying a single bias voltage

applied to the phase shifter. In this design, the varactor capacitance is tuned from 0.3 pF

to 1.2 pF. The inductance values for L1, L2 and L3 are 1.7 nH, 4.7 nH, 7.1 nH. The

resultant phase shift can be tuned from -150 to 150 degrees (Fig. 4.12) while the

reflection loss varies from 1.5 dB to 1.8 dB within the entire tuning range (Fig. 4.13).

4.3.3 Symmetrical array design

A symmetrical phased array is designed based on the phased array design

approach previously described. The block diagram of the proposed phased array is shown

in Fig. 4.14. Series fed arrays usually suffer from frequency scanning. One way to

alleviate this problem is to design a symmetrical array. Therefore, by placing two sub-

arrays together and feeding a symmetrical structure from the center is expected to

enhance the spontaneous bandwidth of the phased array. The phased array is composed of

two identical serially fed sub-arrays where the input signal is equally divided between

them using a 3-dB power divider. The sub-arrays are composed of directional couplers,

amplifying stages, power combiners and phase shifters as shown in Fig. 4.14. The non-

feeding end of each serially fed sub-array is connected to a reflective phase shifter. The

sub-array has been designed according to the approach previously discussed.

4.4 Fabrication and Measurement Results

An eight-element phased array at 2 GHz based on the proposed method is design

and fabricated (Fig. 4.15). The phased array is fabricated on Rogers RO4003C with a

118
dielectric constant of 3.55 and a thickness of 20 mils. The 3-dB hybrid couplers are

designed and implemented on the same substrate. The surface-mount power combiners

SP-2U2+ from Mini-circuits are used to combine the power at each antenna element. The

ERA2+ amplifiers from Mini-circuits are used to implement the fixed-gain stages while

HP MGA-64135 amplifiers are used as the variable gain stages. The gain of this amplifier

can be varied from 1 dB to 13 dB as its bias voltage is tuned from 8.5 to 13 volts. The

varactor diodes used in the design of phase shifter are MA46H120 from MA/COM with a

tunability of 4:1 and a quality factor of 50 at the design frequency. The varactor

capacitance can be tuned from 0.3 pF to 1.2 pF by varying its bias voltage from -0.5 volts

to -15 volts. The lumped inductors used in the design of the phase shifting circuits are

0603CS-1N8XJL, 0603CS-4N7XJL and 0603CS-6N8XJL from Coilcraft with

inductance values of 1.7 nH, 4.7 nH and 7.1 nH and quality factors of 55, 60 and 80

respectively. Each phase shifter is tuned separately with a single bias voltage ranging

between 0.5 to 15 volts. The S-parameters at each of the eight output ports are measured

at various bias voltages and subsequently the corresponding array factors are calculated.

The plot of the array factors shows a good agreement with the simulated results (Fig.

4.16). A Scan angle of 25 degrees is achieved while the side-lobes are approximately 10

dB below the main beam.

119
Fig. 4.15. Photograph of the symmetrical phased array

Fig. 4.16. The normalized array factor based on measured S-parameters compared to the
simulation.

120
4.5 Chapter Conclusion

A new approach to the design of low complexity phased arrays is presented. In

this approach, phase progression in the entire array is controlled by using a single phase

shifters. This approach is a general method that can be applied to any type of phased

array including passive, active or series, parallel in both of receive and transmit mode. A

new phased array based on this approach is designed. The proposed phased array is

composed of couplers, amplifying stages, power combiners and two phase shifters. The

phase shifters are simple and compact consisting of varactor diodes and lumped

inductors. The array scan angle can be controlled with a single bias voltage. In order to

validate the design, an eight-element phased array at 2 GHz is fabricated and measured.

Based on the measured S-parameters for the eight element array, the corresponding array

factor is calculated. A scan angle of 25 degrees is achieved demonstrating a very good

agreement with the simulated results.

121
 Chapter 5

Conclusion and Future Work

In conclusion, this thesis presents several new approaches to the design of phased

arrays with reduced complexity. As mentioned before, phased array antenna systems

have traditionally been in use for military applications in the past several decades.

However, there has also been recent growth in civilian radar-based sensors and advanced

communication systems that have drawn an increasing interest in utilizing phased array

technologies for commercial applications. Phased array, capable of providing a

directional beam that can be electronically steered, can significantly enhance the

performance of sensors and communications systems. The spatial selectivity of phased

arrays can increase the channel capacity and data rate without requiring extra bandwidth.

The directional beam of a phased array also allows for a more efficient power

management. In addition, the spatial filtering nature of the phased array systems alleviate

the problem of multipath fading and co-channel interference by suppressing signals

emanating from undesirable directions. Despite the broad range of potential phased array

applications, phased array technology has not been widely deployed in the commercial

arena. The high cost and complexity of phased arrays is the primary impediment to their

122
deployment in any large-scale commercial application. Thus, any substantial reduction in

the cost and complexity of phased array systems will facilitate their much wider use. This

thesis presents several new approaches to the design of phased arrays which can

significantly reduce their complexity.

5.1 Thesis Summary

In the Chapter 2, a new method to design phased arrays based on the extended

resonance technique is presented. In this method, power dividing and phase shifting tasks

are performed using the same circuitry. Unlike the conventional phased arrays, this

technique eliminates the need for a separate phase shifter per each antenna element hence

reducing their complexity. In the design of the phased array, extended resonance

technique is utilized with heterodyne mixing to achieve uniform power distribution at the

LO/IF paths and to perform phase-shifting at the IF-stage. A new extended resonance

circuit topology has been employed in the IF stage to enhance the scan range of the

phased array. In addition to the design description of the IF-stage, comprehensive

analysis of the maximum achievable phase shift in the new extended resonance circuit is

presented. Furthermore, a modular approach to realize scalable phased array design is

introduced. The design complexity associated with large phased arrays can be

significantly reduced by using the scalable phased array design introduced here. Finally, a

complete modular 7-element 24 GHz has been demonstrated.

Because of the cumulative loss of ladder networks, extended resonance technique

can be used to design phased array modules with 8-12 array elements. Increasing the size

of the phased array beyond this limit can be realized based on the modular approach

123
described in chapter 2. The modular approach allows placing several circuit modules in

tandem to expand the size of the array up to 80-120 elements. Furthermore, the extended

resonance phased arrays can achieve a scan range as wide as +/-90 degrees. Therefore, in

principle, extended resonance technique can be used to design phased arrays with a broad

scan range. The extended resonance phased array can achieve a bandwidth of several

percents in order to minimize the effect of beam squint. Therefore, it would be more

appropriate to utilize extended resonance phased array for applications requiring a

relatively narrow bandwidth.

In Chapter 3, a new bi-directional phased array design is presented. This design

takes advantage of a bi-directional series feed network to reduce the phase tuning

requirement in a phased array. As the result, complexity of the phase shifters required in

the array can be reduced resulting in overall phased array cost and complexity reduction.

A general design procedure for a bidirectional n-element phased array feed network is

presented. Furthermore, a compact phase shifter is designed and utilized in the phased

array. The overall architecture allows one to steer the radiation beam by a single control

voltage obviating the need for complex bias circuits. As a proof-of-principle, an eight-

element bi-directionally fed phased array at 2.4 GHz is designed, fabricated and its

measurement results are reported.

The bi-directional series-fed phased array can be designed to incorporate 10-20

array elements. Increasing the size of the phased array beyond this range would be

possible if phase shifters with low loss were available. The bi-directional phased array

allows scan ranges as wide as +/-90 degrees. This approach is desirable to achieve a wide

124
scan range phase arrays employing low complexity phase shifters. The beam squint

characteristic of the bi-directional phased array depends on the type of the phase shifter

used in the array.

In Chapter 3, a new approach to the design phased arrays with low complexity is

presented. Unlike the conventional design of phased arrays requiring a separate phase

shifter for each antenna element, in the proposed design, the phase progression across the

antenna elements can be controlled by using only a single phase shifter. As a result, the

cost and complexity of the phased arrays can be significantly reduced based on this

design. The proposed phased array is composed of couplers, amplifying stages, power

dividers and two phase shifters. The phase shifters are placed at the two opposite ends of

a serially fed array. Implementation and measurement results for an eight-element phased

array designed based on this approach to operate at 2 GHz are presented.

Considering the dependence of achievable scan range on the number of antenna

elements, the approach above can be used to design phased array modules having of 5-10

elements. Increasing the size of the phased array beyond this limit can be realized based

on the modular approach described in chapter 2. The modular approach allows for

placing several circuit modules in tandem to expand the size of the array up to 20-40

elements. The frequency performance of the proposed phased array depends on the type

of the feed network used in the array. Parallel feed network can be used to realize phased

arrays based on this approach to achieve a wide bandwidth performance.

In the first chapter, different techniques to reduce the complexity of phased arrays

by reducing the number of required phase shifters were discussed. Extended resonance

125
technique does not need separate power dividing and phase shifting circuits as compared

to sub-arrayed antenna approaches. Furthermore, it can be used at RF stage, eliminating

the need for LO signal distribution circuits and the accompanying large number of

mixers. Like any other series-fed arrays, the extended resonance phased array may show

a higher beam squint compared to the other approaches discussed here. Since the losses

of passive elements are cumulative in a ladder network and, the performance of the arrays

designed based on these approaches are more prone to losses as the number of antenna

elements increases. The modular approach presented in the second chapter can be

exploited to alleviate this problem. The novel phased array design approach described in

chapter 4 uses a substantially lower number of phase shifters while it doesn’t suffer from

increase of side-lobe levels and emergence of grating lobes that are usually seen in sub-

arrayed antenna arrays.

5.2 Future Work

5.2.1 Extended Resonance Phased Array

Extended resonance can significantly reduce the complexity of phased arrays by

achieving phase shifting and power division in a single network. Extended resonance can

be utilized at RF stage to realize a passive phased array. Passive phased arrays have

advantage of high dynamic range, better power handling and higher linearity. In this

phased array, extended resonance will be used to directly phase shift and combine the

signals from antenna elements. In this case, there will be no need for several mixers and

active elements. In order to accomplish this at mm-wave, varactors will high quality

factor are required. MEMS varactors have shown to achieve high quality factors at this

126
frequency regime. Furthermore, as shown in chapter 2, increasing the capacitance value

of varactors can increase the maximum achievable inter-element phase shift in extended

resonance circuit. MEMS varactors can usually provide higher capacitance values

compared to the other conventional varactor technologies. Therefore using MEMS

varactors allows extended resonance to operate at RF stage resulting in entirely passive

phased array.

A modular approach has been presented in chapter 2 to realize a scalable extended

resonance phased array. However, the presented approach has been used for transmit

phased array. Extended resonance circuit can be modified to apply the modular approach

to receive phased array as well. The modular approach can significantly help to achieve

large size phased arrays as the number of antenna elements increases; the loss in the

network accumulates and it can negatively affect the uniform power distribution across

the array.

5.2.2 Bi-directional Series-Fed Phased Array

Similar to any series-fed phased array, the loss of phase shifters and couplers can

accumulate across the bi-directional array presented in chapter 3. The bi-directional

feeding has already shown to reduce the number of tuning elements in the required phase

shifter. Therefore, the phase shifters required in this phased array demonstrate relatively

lower loss due to the reduced number of tuning elements. This allows the bidirectional

feeding method to be applied to phased arrays implemented using integrated circuits

where the loss of lumped element is relatively higher compared to the discrete elements.

In integrated circuit, the lumped elements especially inductors occupy the major die area,

127
therefore reducing the number of lumped elements can substantially reduce the die size

and thus the overall cost of the system. The bidirectional feeding approach also improve

the performance of the phased array by reducing the lumped elements in the integrated

phased arrays as quality of lumped element is usually a major limitation in the design of

tunable integrated circuits.

5.2.3 Phased Array Using a Single Phase Shifter

In order to take advantage of the unique capabilities that phased array technology

is offering for the recent commercial applications, certain strict requirements must be

met. The new generation of phased arrays must be small, power efficient and

inexpensive. Such phased arrays can then be integrated into the existing electronic

appliances such as the iPhone and laptop computers. Recent advances in silicon based

device performance accompanied with the high density of circuit integration have

provided a unique opportunity to drastically lower the cost of phased arrays by

implementing the entire phased array on a single chip. Furthermore, integration of all the

required circuit modules on a silicon chip obviates the need for any external interconnects

resulting in a significant performance improvement.

Ideally, a tunable true time delay that provides frequency independent operation is

needed for each array element to prevent beam squint and bandwidth reduction.

However, integration of high performance broadband and low loss tunable true time

delays for each antenna element again results in excessively high power consumption and

large die area requirement. Furthermore, each of the tunable time delays usually requires

several independently controlled bias voltages. Thus, the design of true time delay based

128
phased arrays with a large number of array elements is very challenging due to the

exceedingly large number of control voltages and their associated complex circuitry. On

the other hand, for instance, in the next generation of gigabit rate commercial

communication systems at 60 GHz, transmit power levels of up to 39 dBm equivalent

isotropic radiated power (EIRP) are allowed. Considering the maximum power rating of

power amplifiers at millimeter wave region, tens of array elements are required to

generate such transmit powers levels. In order to realize larger phased arrays, the current

integrated arrays employ compact but narrowband RF phase shifting architectures

(narrow-bandwidth approximation of the true time delay). At the same time, the

unlicensed band at 60 GHz is offering bandwidths as high as 7 GHz. Clearly, the full

potential of this frequency band for the development of the next generation personal

communication systems providing the highest possible data rates will not be fully

realized unless the available bandwidth is efficiently exploited.

The phased array design approach presented in chapter 3 allows using a single

phase shifter in the entire phased array; therefore, true tunable time delay can be used in

this phase array to achieve high bandwidth. As result, phased arrays integrated on a single

chip can be realized by this method to achieve performance that has not been possible

due to large number of phase shifters required in conventional phased arrays. The

integrated phased arrays with high performance such as wide operating bandwidth play

key role for the emerging 60 GHz short range wireless systems.

129
BIBLIOGRAPHY

130
 BIBLIOGRAPHY

[1] G. Marconi, "Directive Antenna," Proc. Royal Soc, vol. 77A, 1906.

[2] Wikipedia. Karl Ferdinand Braun. Available:

http://en.wikipedia.org/wiki/Karl_Ferdinand_Braun

[3] C. W. R. a. E. W. K. H. H. Beverage, "The wave antenna," J. Am. Inst. Electr.

Eng., vol. XLII, p. 12, March 1923.

[4] R. J. Mailloux, "Electronically scanned arrays," Synthesis Lectures on Antennas,

vol. 2, pp. 1-82, 2007.

[5] H. T. Friis and C. B. Feldman, "A Multiple Unit Steerable Antenna for Short-
Wave Reception," Proceedings of the Institute of Radio Engineers, vol. 25, pp.
841-917, 1937.

[6] A. G. Fox, "An Adjustable Wave-Guide Phase Changer," Proceedings of the IRE,
vol. 35, pp. 1489-1498, 1947.

[7] The Nobel Prize in Physics 1968. Available:

http://nobelprize.org/nobel_prizes/physics/laureates/1968/alvarez-bio.html

[8] Wikipedia. Phased Array. Available: http://en.wikipedia.org/wiki/Phased_array

[9] E. Brookner, "Phased-array radars," Scientific American, vol. 252, pp. 94-102,
1985.

[10] J. Wenger, "Automotive radar - status and perspectives," in Compound

Semiconductor Integrated Circuit Symposium, 2005. CSIC '05. IEEE, 2005, p. 4
pp.

[11] I. Gresham, A. Jenkins, R. Egri, C. Eswarappa, F. Kolak, R. Wohlert, J. Bennett,

and J. P. Lanteri, "Ultra wide band 24GHz automotive radar front-end," in
Microwave Symposium Digest, 2003 IEEE MTT-S International, 2003, pp. 369-
372 vol.1.

[12] R. Schneider and J. Wenger, "System aspects for future automotive radar," in
Microwave Symposium Digest, 1999 IEEE MTT-S International, 1999, pp. 293-
296 vol.1.

[13] W. D. Jones, "Keeping cars from crashing," IEEE Spectrum, p. 46, Sep 2001.

131
[14] A. Valdes-Garcia, S. Nicolson, L. Jie-Wei, A. Natarajan, C. Ping-Yu, S.
Reynolds, J. H. C. Zhan, and B. Floyd, "A SiGe BiCMOS 16-element phased-
array transmitter for 60GHz communications," in Solid-State Circuits Conference
Digest of Technical Papers (ISSCC), 2010 IEEE International, 2010, pp. 218-
219.

[15] S. K. Reynolds, A. S. Natarajan, T. Ming-Da, S. Nicolson, Z. Jing-Hong Conan,

L. Duixian, D. G. Kam, O. Huang, A. Valdes-Garcia, and B. A. Floyd, "A 16-
element phased-array receiver IC for 60-GHz communications in SiGe
BiCMOS," in Radio Frequency Integrated Circuits Symposium (RFIC), 2010
IEEE, 2010, pp. 461-464.

[16] S. K. Reynolds, B. A. Floyd, U. R. Pfeiffer, T. Beukema, J. Grzyb, C. Haymes, B.

Gaucher, and M. Soyuer, "A Silicon 60-GHz Receiver and Transmitter Chipset
for Broadband Communications," Solid-State Circuits, IEEE Journal of, vol. 41,
pp. 2820-2831, 2006.

[17] E. Cohen, C. G. Jakobson, S. Ravid, and D. Ritter, "A Bidirectional TX/RX Four-
Element Phased Array at 60 GHz With RF-IF Conversion Block in 90-nm CMOS
Process," Microwave Theory and Techniques, IEEE Transactions on, vol. 58, pp.
1438-1446, 2010.

[18] N. Guo, R. C. Qiu, S. S. Mo, and K. Takahashi, "60-GHz millimeter-wave radio:

principle, technology, and new Results," EURASIP Journal on Wireless
Communications and Networking, vol. 2007, pp. 48-48, 2007.

[19] C. Alakija and S. P. Stapleton, "A mobile base station phased array antenna," in
Wireless Communications, 1992. Conference Proceedings., 1992 IEEE
International Conference on Selected Topics in, 1992, pp. 118-121.

[20] P. V. Brennan, "Low cost phased array antenna for land-mobile satcom
applications," Microwaves, Antennas and Propagation, IEE Proceedings H, vol.
138, pp. 131-136, 1991.

[21] P. F. Turner, "Regional Hyperthermia with an Annular Phased Array,"

Biomedical Engineering, IEEE Transactions on, vol. BME-31, pp. 106-114, 1984.

[22] M. O'Donnell, "Efficient parallel receive beam forming for phased array imaging
using phase rotation [medical US application]," in Ultrasonics Symposium, 1990.
Proceedings., IEEE 1990, 1990, pp. 1495-1498 vol.3.

[23] J. Souquet, P. Hanrath, L. Zitelli, P. Kremer, B. A. Langenstein, and M. Schluter,

"Transesophageal Phased Array for Imaging the Heart," Biomedical Engineering,
IEEE Transactions on, vol. BME-29, pp. 707-712, 1982.

132
[24] L. Wang, H. Hricak, M. W. Kattan, L. H. Schwartz, S. C. Eberhardt, H. N. Chen,
and P. T. Scardino, "Combined endorectal and phased-array MRI in the prediction
of pelvic lymph node metastasis in prostate cancer," American Journal of
Roentgenology, vol. 186, p. 743, 2006.

[25] A. Villers, P. Puech, D. Mouton, X. Leroy, C. Ballereau, and L. Lemaitre,

"Dynamic contrast enhanced, pelvic phased array magnetic resonance imaging of
localized prostate cancer for predicting tumor volume: correlation with radical
prostatectomy findings," The Journal of urology, vol. 176, pp. 2432-2437, 2006.

[26] L. Qian, L. Qingming, and B. Chance, "2D phased array fluorescence wireless
localizer in breast cancer detection," in Computer Architectures for Machine
Perception, 2003 IEEE International Workshop on, 2004, pp. 71-73.

[27] D. Parker and D. C. Zimmermann, "Phased arrays - part 1: theory and

architectures," Microwave Theory and Techniques, IEEE Transactions on, vol.
50, pp. 678-687, 2002.

[28] J. H. Winters, "Smart antennas for wireless systems," Personal Communications,

IEEE, vol. 5, pp. 23-27, 1998.

[29] Y. Lo and S. Lee, Antenna Handbook: Applications: Kluwer Academic

Publishers, 1993.

[30] R. C. Hansen, Phased array antennas: Wiley-Interscience, 2009.

[31] R. C. Johnson and H. Jasik, "Antenna engineering handbook," 1984.

[32] R. J. Mailloux and I. Books, Phased array antenna handbook: Artech House,
2005.

[33] D. M. Pozar and D. H. Schaubert, "Comparison of three series fed microstrip

array geometries," in Antennas and Propagation Society International
Symposium, 1993. AP-S. Digest, 1993, pp. 728-731 vol.2.

[34] M. A. Y. Abdalla, K. Phang, and G. V. Eleftheriades, "A Steerable Series-fed

Phased Array Architecture Using Tunable PRI/NRI Phase Shifters," in Antenna
Technology: Small Antennas and Novel Metamaterials, 2008. iWAT 2008.
International Workshop on, 2008, pp. 83-86.

[35] E. Cohen, C. Jakobson, S. Ravid, and D. Ritter, "A bidirectional TX/RX four
element phased-array at 60GHz with RF-IF conversion block in 90nm CMOS
process," in Radio Frequency Integrated Circuits Symposium, 2009. RFIC 2009.
IEEE, 2009, pp. 207-210.

133
[36] K. Jeong-Geun, K. Dong-Woo, M. Byung-Wook, and G. M. Rebeiz, "A single-
chip 36-38 GHz 4-element transmit/receive phased-array with 5-bit amplitude and
phase control," in Microwave Symposium Digest, 2009. MTT '09. IEEE MTT-S
International, 2009, pp. 561-564.

[37] K. Kwang-Jin and G. M. Rebeiz, "0.13um CMOS Phase Shifters for X-, K-, and
K-Band Phased Arrays," Solid-State Circuits, IEEE Journal of, vol. 42, pp. 2535-
2546, 2007.

[38] A. Natarajan, S. K. Reynolds, T. Ming-Da, S. T. Nicolson, J. H. C. Zhan, K. Dong

Gun, L. Duixian, Y. L. O. Huang, A. Valdes-Garcia, and B. A. Floyd, "A Fully-
Integrated 16-Element Phased-Array Receiver in SiGe BiCMOS for 60-GHz
Communications," Solid-State Circuits, IEEE Journal of, vol. 46, pp. 1059-1075,
2011.

[39] A. M. Niknejad and H. Hashemi, mm-Wave silicon technology: 60GHz and

beyond: Springer Verlag, 2008.

[40] K. Dong-Woo, L. Hui Dong, K. Chung-Hwan, and H. Songcheol, "Ku-band

MMIC phase shifter using a parallel resonator with 0.18-/spl mu/m CMOS
technology," Microwave Theory and Techniques, IEEE Transactions on, vol. 54,
pp. 294-301, 2006.

[41] H. Zarei, C. T. Charles, and D. J. Allstot, "Reflective-Type Phase Shifters for

Multiple-Antenna Transceivers," Circuits and Systems I: Regular Papers, IEEE
Transactions on, vol. 54, pp. 1647-1656, 2007.

[42] M. Byung-Wook and G. M. Rebeiz, "Single-Ended and Differential Ka-Band

BiCMOS Phased Array Front-Ends," Solid-State Circuits, IEEE Journal of, vol.
43, pp. 2239-2250, 2008.

[43] Y. Tiku and G. M. Rebeiz, "A 22-24 GHz 4-Element CMOS Phased Array With
On-Chip Coupling Characterization," Solid-State Circuits, IEEE Journal of, vol.
43, pp. 2134-2143, 2008.

[44] W. Chao-Shiun, H. Juin-Wei, C. Kun-Da, and W. Chorng-Kuang, "A 60-GHz

Phased Array Receiver Front-End in 0.13-um CMOS Technology," Circuits and
Systems I: Regular Papers, IEEE Transactions on, vol. 56, pp. 2341-2352, 2009.

[45] K. Scheir, S. Bronckers, J. Borremans, P. Wambacq, and Y. Rolain, "A 52 GHz

Phased-Array Receiver Front-End in 90 nm Digital CMOS," Solid-State Circuits,
IEEE Journal of, vol. 43, pp. 2651-2659, 2008.

[46] A. Natarajan, A. Komijani, and A. Hajimiri, "A fully integrated 24-GHz phased-
array transmitter in CMOS," Solid-State Circuits, IEEE Journal of, vol. 40, pp.
2502-2514, 2005.

134
[47] D. Ehyaie and A. Mortazawi, "A 24 GHz modular phased array based on
extended resonance technique," in Phased Array Systems and Technology
(ARRAY), 2010 IEEE International Symposium on, 2010, pp. 814-818.

[48] D. Ehyaie and A. Mortazawi, "A 24-GHz Modular Transmit Phased Array,"
Microwave Theory and Techniques, IEEE Transactions on, vol. 59, pp. 1665-
1672, 2011.

[49] J. D. Fredrick, W. Yuanxun, and T. Itoh, "A smart antenna receiver array using a
single RF channel and digital beamforming," Microwave Theory and Techniques,
IEEE Transactions on, vol. 50, pp. 3052-3058, 2002.

[50] I. Chiba, R. Miura, T. Tanaka, and Y. Karasawa, "Digital beam forming (DBF)
antenna system for mobile communications," Aerospace and Electronic Systems
Magazine, IEEE, vol. 12, pp. 31-41, 1997.

[51] B. C. Kane, L. A. Geis, M. A. Wyatt, D. G. Copeland, and J. A. Mogensen,

"Smart Phased Array SoCs: A Novel Application for Advanced SiGe HBT
BiCMOS Technology," Proceedings of the IEEE, vol. 93, pp. 1656-1668, 2005.

[52] A. Tombak and A. Mortazawi, "A novel low-cost beam-steering technique based
on the extended-resonance power-dividing method," Microwave Theory and
Techniques, IEEE Transactions on, vol. 52, pp. 664-670, 2004.

[53] J. T. Nemit, "NETWORK APPROACH FOR REDUCING THE," ed: Google

Patents, 1974.

[54] A. Abbaspour-Tamijani and K. Sarabandi, "An affordable millimeter-wave beam-

steerable antenna using interleaved planar subarrays," Antennas and Propagation,
IEEE Transactions on, vol. 51, pp. 2193-2202, 2003.

[55] W. T. Patton, "Limited scan arrays," Proceedings of Phased Array Antenna

Symposium, pp. 332-343, 1972.

[56] S. P. Skobelev, "Methods of constructing optimum phased-array antennas for

limited field of view," Antennas and Propagation Magazine, IEEE, vol. 40, pp.
39-50, 1998.

[57] R. Mailloux, L. Zahn, and A. Martinez, III, "Grating lobe control in limited scan
arrays," Antennas and Propagation, IEEE Transactions on, vol. 27, pp. 79-85,
1979.

[58] J. Howell, "Limited scan antennas," in Antennas and Propagation Society

International Symposium, 1974, 1974, pp. 117-120.

135
[59] S. Sugiura and H. Iizuka, "Reactively Steered Ring Antenna Array for
Automotive Application," Antennas and Propagation, IEEE Transactions on, vol.
55, pp. 1902-1908, 2007.

[60] S. Chen, A. Hirata, T. Ohira, and N. C. Karmakar, "Fast beamforming of

electronically steerable parasitic array radiator antennas: theory and experiment,"
Antennas and Propagation, IEEE Transactions on, vol. 52, pp. 1819-1832, 2004.

[61] R. Harrington, "Reactively controlled directive arrays," Antennas and

Propagation, IEEE Transactions on, vol. 26, pp. 390-395, 1978.

[62] K. Aamo, "Frequency controlled antenna beam steering," in Microwave

Symposium Digest, 1994., IEEE MTT-S International, 1994, pp. 1549-1552 vol.3.

[63] M. Kim, J. B. Hacker, A. L. Sailer, and J. H. Hong, "A heterodyne-scan phased-

array antenna," Microwave and Guided Wave Letters, IEEE, vol. 9, pp. 535-537,
1999.

[64] T. Nishio, X. Hao, W. Yuanxun, and T. Itoh, "A frequency-controlled active

phased array," Microwave and Wireless Components Letters, IEEE, vol. 14, pp.
115-117, 2004.

[65] J. F. Buckwalter, A. Babakhani, A. Komijani, and A. Hajimiri, "An Integrated

Subharmonic Coupled-Oscillator Scheme for a 60-GHz Phased-Array
Transmitter," Microwave Theory and Techniques, IEEE Transactions on, vol. 54,
pp. 4271-4280, 2006.

[66] K. D. Stephan, "Inter-Injection-Locked Oscillators for Power Combining and

Phased Arrays," Microwave Theory and Techniques, IEEE Transactions on, vol.
34, pp. 1017-1025, 1986.

[67] R. A. York and T. Itoh, "Injection- and phase-locking techniques for beam control
[antenna arrays]," Microwave Theory and Techniques, IEEE Transactions on, vol.
46, pp. 1920-1929, 1998.

[68] R. Adler, "A Study of Locking Phenomena in Oscillators," Proceedings of the

IRE, vol. 34, pp. 351-357, 1946.

[69] A. L. Martin and A. Mortazawi, "A class-E power amplifier based on an extended
resonance technique," Microwave Theory and Techniques, IEEE Transactions on,
vol. 48, pp. 93-97, 2000.

[70] A. L. Martin, A. Mortazawi, and B. C. Deloach, Jr., "An eight-device extended-

resonance power-combining amplifier," Microwave Theory and Techniques, IEEE
Transactions on, vol. 46, pp. 844-850, 1998.

136
[71] A. L. Martin and A. Mortazawi, "A new lumped-elements power-combining
amplifier based on an extended resonance technique," Microwave Theory and
Techniques, IEEE Transactions on, vol. 48, pp. 1505-1515, 2000.

[72] A. Mortazawi and B. C. de Leach, Jr., "Spatial power combining oscillators based
on an extended resonance technique," Microwave Theory and Techniques, IEEE
Transactions on, vol. 42, pp. 2222-2228, 1994.

[73] A. Martin, A. Mortazawi, and B. C. De Loach, Jr., "A power amplifier based on
an extended resonance technique," Microwave and Guided Wave Letters, IEEE,
vol. 5, pp. 329-331, 1995.

[74] A. Tombak and A. Mortazawi, "An X-band low-cost phased array based on the
extended resonance power dividing technique," in Antennas and Propagation
Society International Symposium, 2005 IEEE, 2005, pp. 334-337 Vol. 1A.

[75] A. Tombak and A. Mortazawi, "A novel phased array based on the extended
resonance power dividing technique," in Microwave Symposium Digest, 2003
IEEE MTT-S International, 2003, pp. 1849-1852 vol.3.

[76] A. Tombak and A. Mortazawi, "Design of novel low-cost phased arrays based on
the extended resonance technique," in Antennas and Propagation Society
International Symposium, 2003. IEEE, 2003, pp. 672-675 vol.4.

[77] S. W. Alland, "Antenna requirements and architecture tradeoffs for an automotive

forward looking radar," in Radar Conference, 1998. RADARCON 98.
Proceedings of the 1998 IEEE, 1998, pp. 367-372.

[78] C. A. Balanis, Antenna theory: Wiley New York, 1997.

[79] M. Tsuji, T. Nishikawa, K. Wakino, and T. Kitazawa, "Bi-directionally fed

phased-array antenna downsized with variable impedance phase shifter for ISM
band," Microwave Theory and Techniques, IEEE Transactions on, vol. 54, pp.
2962-2969, 2006.

[80] E. Ojefors, C. Shi, K. From, I. Skarin, P. Hallbjorner, and A. Rydberg,

"Electrically Steerable Single-Layer Microstrip Traveling Wave Antenna With
Varactor Diode Based Phase Shifters," Antennas and Propagation, IEEE
Transactions on, vol. 55, pp. 2451-2460, 2007.

[81] C. Shi, E. Ojefors, P. Hallbjorner, and A. Rydberg, "Compact reflective

microstrip phase shifter for traveling wave antenna applications," Microwave and
Wireless Components Letters, IEEE, vol. 16, pp. 431-433, 2006.

[82] L. Sungjoon, C. Caloz, and T. Itoh, "Metamaterial-based electronically controlled

transmission-line structure as a novel leaky-wave antenna with tunable radiation

137
 angle and beamwidth," Microwave Theory and Techniques, IEEE Transactions
on, vol. 52, pp. 2678-2690, 2004.

[83] M. A. Y. Abdalla, K. Phang, and G. V. Eleftheriades, "A Planar Electronically

Steerable Patch Array Using Tunable PRI/NRI Phase Shifters," Microwave
Theory and Techniques, IEEE Transactions on, vol. 57, pp. 531-541, 2009.

[84] E. Brookner, Practical phased-array antenna systems: Artech house, 1991.

[85] D. M. Pozar and B. Kaufman, "Comparison of three methods for the measurement
of printed antenna efficiency," Antennas and Propagation, IEEE Transactions on,
vol. 36, pp. 136-139, 1988.

[86] P. Hallbjorner, M. Bergstrom, M. Boman, P. Lindberg, E. Ojefors, and A.

Rydberg, "Millimetre-wave switched beam antenna using multiple travelling-
wave patch arrays," Microwaves, Antennas and Propagation, IEE Proceedings -,
pp. 551-555, 2005.

[87] A. Natarajan, B. Floyd, and A. Hajimiri, "A Bidirectional RF-Combining 60GHz

Phased-Array Front-End," in Solid-State Circuits Conference, 2007. ISSCC 2007.
Digest of Technical Papers. IEEE International, 2007, pp. 202-597.

[88] F. Ellinger, U. Mayer, M. Wickert, N. Joram, J. Wagner, R. Eickhoff, I.

Santamaria, C. Scheytt, and R. Kraemer, "Integrated Adjustable Phase Shifters,"
Microwave Magazine, IEEE, vol. 11, pp. 97-108, 2010.

[89] N. S. Barker and G. M. Rebeiz, "Optimization of distributed MEMS transmission-

line phase shifters-U-band and W-band designs," Microwave Theory and
Techniques, IEEE Transactions on, vol. 48, pp. 1957-1966, 2000.

[90] A. Abbaspour-Tamijani, L. Dussopt, and G. M. Rebeiz, "Miniature and tunable

filters using MEMS capacitors," Microwave Theory and Techniques, IEEE
Transactions on, vol. 51, pp. 1878-1885, 2003.

[91] J. S. Hayden and G. M. Rebeiz, "Low-loss cascadable MEMS distributed X-band

phase shifters," Microwave and Guided Wave Letters, IEEE, vol. 10, pp. 142-144,
2000.

[92] E. G. Erker, A. S. Nagra, Y. Liu, P. Periaswamy, T. R. Taylor, J. Speck, and R. A.

York, "Monolithic Ka-band phase shifter using voltage tunable BaSrTiO3 parallel
plate capacitors," Microwave and Guided Wave Letters, IEEE, vol. 10, pp. 10-12,
2000.

[93] B. Acikel, T. R. Taylor, P. J. Hansen, J. S. Speck, and R. A. York, "A new high
performance phase shifter using BaxSr1-x TiO3 thin films," Microwave and
Wireless Components Letters, IEEE, vol. 12, pp. 237-239, 2002.

138
[94] D. Kuylenstierna, A. Vorobiev, P. Linnér, and S. Gevorgian, "Composite
right/left handed transmission line phase shifter using ferroelectric varactors,"
Microwave and Wireless Components Letters, IEEE, vol. 16, pp. 167-169, 2006.

[95] A. S. Nagra, J. Xu, E. Erker, and R. A. York, "Monolithic GaAs phase shifter
circuit with low insertion loss and continuous 0-360 phase shift at 20 GHz,"
Microwave and Guided Wave Letters, IEEE, vol. 9, pp. 31-33, 1999.

[96] A. S. Nagra and R. A. York, "Distributed analog phase shifters with low insertion
loss," Microwave Theory and Techniques, IEEE Transactions on, vol. 47, pp.
1705-1711, 1999.

[97] F. Ellinger, H. Jackel, and W. Bachtold, "Varactor-loaded transmission-line phase

shifter at C-band using lumped elements," Microwave Theory and Techniques,
IEEE Transactions on, vol. 51, pp. 1135-1140, 2003.

[98] D. Ehyaie and A. Mortazawi, "A modular extended resonance transmit phased
array with improved scan angle," 2009, pp. 1557-1560.

139