Piotr Kowina

RF Measurements Techniques Christine Völlinger

Manfred Wendt

2023
RF for Accelerators

Backround Information

The CERN Accelerator School

in collaboration with

The Helmholz-Zentrum Berlin (HZB)

Berlin - Germany
18th Juni - 1st July 2023
RF Measurement Concepts

P. Kowina2 , C. Völlinger1 and M. Wendt1

1 CERN, Geneva, Switzerland
2 GSI, Darmstadt, Germany

Abstract
For the characterization of components, systems and signals in the range of
microwave and radio-frequencies (RF) specific equipment and dedicated mea-
surement instruments are used. In this article the fundamentals of RF signal
processing and measurement techniques are discussed. While this document
was not updated in recent years, it still provides valuable, complementary
background information for the introduction to the RF Measurement Tech-
niques lecture and the Practical RF “Hands-on” Course, which are part of
the Advanced Accelerator Physics training program and the Special Topic RF
Course of the CERN Accelerator School (CAS).
A key element in the so-called RF front end of almost every RF instruments
is the Schottky diode, which is used for both, as RF mixer, but also as sam-
pling switch for the analog signal. The spectrum analyzer (SA) has become
an absolutely indispensable tool for the analysis of RF and microwave signals,
which also can be beam related signals. The SA front end is a RF downcon-
verter, in modern spectrum analyzers often a rather complex architecture of up-
and downconverters. Some of the reasons for this complexity, certain working
principles, as well as limitations are discussed. In addition, an overview of
the development of scalar and vector signal analyzers are given. The defini-
tion of the noise temperature for a one-port, and the noise figure of a two-port
RF system are presented along with basic concepts and relations, as well as
a brief discussion of commonly used noise-measurement techniques. In the
following we discuss the operating principles of the network analyzer, and
explain the differences between scalar and vector network analyzers (VNA),
and their methods to measure the transmission and/or reflection coefficients
of the scattering parameters (S-parameter). With the rise of affordable dig-
ital signal processing techniques in recent years, these instruments are now
very versatile and powerful. The (inverse) Fourier transformation capability
adds time-domain measurement options to the VNA, and even allows the re-
moval of undesired parts of the signal trace by time-domain gating. Network
analyzers require sophisticated calibration procedures, which are mandatory
for most measurement applications. Non-linear network analysis methods are
performed to complete the characterization of RF amplifiers using a modern
VNA. The Smith chart is a very valuable and important tool that facilitates
the interpretation of S-parameter measurements. The last part of this article
gives a brief overview how to use the chart, it’s definition, as well as an in-
troduction how to navigate inside the chart are illustrated. Typical examples
show the large variety and flexible usage of the Smith chart for very different
applications and RF problems.
1 A note to the history of RF signal receiving and measurement techniques
In the early days of radio-frequency (RF) engineering the available instrumentation for measurements
was rather limited. Besides elements acting on the heat developed by RF power (bi-metal contacts
and resistors with a very high temperature coefficient) only point/contact diodes, and to some extent
vacuum tubes, were available as signal detectors. For several decades the slotted measurement line,
see Section 8.1, was the only commonly used instrument to measure impedances and complex reflection
coefficients. Around 1960 the tedious work with these coaxial and waveguide measurement lines became
considerably simplified with the availability of the vector network analyzer. At the same time the first
sampling oscilloscopes with 1 GHz bandwidth arrived on the market. This was possible due to progress in
solid-state (semiconductor) technology and advances in microwave elements (microstrip lines). Reliable,
stable and easily controllable microwave sources are the backbone of spectrum and network analyzers,
as well as sensitive (low-noise) receivers. The following sections focus on signal receiving devices such
as spectrum analyzers. An overview of network analysis is given later in Section 5.

2 Basic definitions, elements and concepts

Before discussing key RF measurement devices, a brief overview of the most important components used
in these devices and the related basic concepts are presented.

2.1 Decibel
Since the unit decibel (dB) is frequently used in RF engineering, a short introduction and definition of
the terms are given. The decibel is a unit used to express relative differences between quantities, e.g. of
signal power. It is expressed as the base-10 logarithm of the ratio of the powers between two signals:

P [dB] = 10 · log(P/P0 ). (1)

It is also common to express the signal amplitude in dB. Since power is proportional to the square of the
signal amplitude, a voltage ratio in dB is expressed as:

V [dB] = 20 · log(V /V0 ). (2)

In Eqs. (1) and (2), P0 and V0 are the reference power and voltage, respectively. A given value in dB
is the same for power ratios as for voltage ratios. It is important to note that there are no ‘power dB’
or ‘voltage dB’ as dB values always express a ratio. Conversely, the absolute power and voltage can be
obtained from dB values by
P [dB]
P = P0 · 10 10 , (3)
V [dB]
V = V0 · 10 20 . (4)

The advantage using a logarithmic scale as unit of the measurement is twofold:

i) typical RF signal powers tends to span several orders of magnitude; and

ii) signal attenuation losses and gains can simply computed by subtraction and addition.

Table 1 helps to familiarize with signal ratios and the associated dB values.
Absolute levels are expressed using a specific reference value, these dB systems are not based
on SI units. Strictly speaking, the reference value should be included in parentheses when giving a dB
value, e.g. +3 dB (1 W) indicates 3 dB at P0 = 1 W, thus 2 W. However, it is more common to add some
typical reference values as letters after the unit, e.g. dBm defines dB using a reference level of P0 = 1
mW. Thus, 0 dBm correspond to −30 dBW, where dBW indicates a reference level of P0 = 1 W. Often
a reference impedance of 50 Ω is assumed. Other common units are:

2
 Table 1: Overview of common dB values and their conversion into power and voltage ratios

Power ratio Voltage ratio

−20 dB 0.01 0.1
−10 dB 0.1 0.32
−6 dB 0.25 0.5
−3 dB 0.50 0.71
−1 dB 0.74 0.89
0 dB 1 1
1 dB 1.26 1.12
3 dB 2.00 1.41
6 dB 4 2
10 dB 10 3.16
20 dB 100 10
n · 10 dB 10n 10n/2

i) dBmV for small voltages with V0 = 1 mV; and

ii) dBmV/m for the electric field strength radiated from an antenna with reference field strength E0 =
1 mV/m.

diode impedance

RF bypass
RF in 50 V capacitor Video out

Fig. 1: Simplified equivalent circuit of a diode detector (w/o parasitic elements)

2.2 The RF diode

One of the most important elements, even today inside the most sophisticated RF measurement devices
is the fast RF diode or Schottky diode. The basic metal–semiconductor junction has an intrinsically very
fast switching time of well below a picosecond, provided that the geometric size and hence the junction
capacitance of the diode has sufficiently small dimensions. However, the unavoidable, and voltage-
dependent junction capacity will lead to limitations of the maximum operating frequency. The simplified
equivalent circuit of such a diode is depicted in Fig. 1 and an example of a commonly used Schottky diode
is shown in Fig. 2. of the most important properties of any diode is its IV-characteristic, which is the
relation of the current passing the diode as a function of the applied voltage [1]. This relation is depicted
graphically for two different types of diodes in Fig. 3. It shows, the diode is a non-ideal commutator
(in contrary to that shown in Fig. 4) for small signals. Note that it is not possible to apply large signals,
since this kind of diode would burn out. Although there exist versions with rather large power handling
capability of Schottky diodes, these can stand more than 9 kV and several tens of amperes, they are not

3
Fig. 2: A typical Schottky diode. The RF input of this detector diode is on the left and the video output
on the right (courtesy Agilent).

I Typical
50 µA/div
LBSD
Typical
Schottky
Diode

V
50 mV/div

Fig. 3: Current as a function of voltage for different diode types (LBSD = low barrier Schottky diode)

Current

Threshold voltage

Voltage
Fig. 4: The current–voltage relation of an ideal commutator with threshold voltage

suitable in microwave applications due to their large junction capacity. The region where the output
voltage is proportional to the input power is called the square-law region (Fig. 5). In this region the input
power is proportional to the square of the input voltage and the output signal is proportional to the input
power, hence the name square-law region.
The transition between the linear region and the square-law region is typically between −10 and
−20 dBm (Fig. 5). For a more detailed description, see [2].
There are some fundamental limitations when using diodes as detectors. The output signal of a
diode (essentially DC or modulated DC if the RF is amplitude modulated) does not contain any phase
information. In addition, the sensitivity of a diode limits the input level range to about −60 dB at best,

4
 500

50

Output voltage [mV]

LBSD
5.0
LBSD

0.5
without load square law loaded

0.05

0.005
-50 -40 -30 -20 -10 0
Input power [dBm]

Fig. 5: Relation between input power and output voltage

which is not sufficient for many applications.

The minimum detectable power level of a RF diode is specified by the ‘tangential sensitivity’,
which typically amounts to −50 to −55 dBm for 10 MHz video bandwidth at the detector output [3].
To overcome these limitations, a more sophisticated method to utilize the RF diode is required.
This method is presented in the next section.

2.3 Mixer
To include the detection of very small RF signals a device with a linear response over a wide range
of signal levels (from 0 dBm (= 1 mW) down to the thermal noise = −174 dBm/Hz = 4·10−21 W/Hz)
is highly preferred. A RF mixer provides these features by using one, two or four diodes in different
configurations (Fig. 6). A mixer is essentially a frequency multiplier with a very high dynamic range,
implementing in it’s simplest form the function

f1 (t) · f2 (t) with f1 (t) = RF signal and f2 (t) = local oscillator (LO) signal (5)

or more explicitly, for two sinusoidal signals with amplitudes ai and frequencies fi (i = 1, 2),
1
a1 cos(2πf1 t + φ) · a2 cos(2πf2 t) = a1 a2 [cos((f1 + f2 )t + φ) + cos((f1 − f2 )t + φ)] . (6)
2
Thus, we obtain a response at the intermediate-frequency (IF) port as sum and difference frequencies of
the local oscillator (LO = f1 ) and RF (= f2 ) signals. Examples of different mixer configurations are
shown in Fig. 6, they all use diodes to multiply the two applied signals, RF and LO. These diodes operate
like a switch, controlled by the frequency of the LO signal (Fig. 7). The response of a mixer in the time
domain is depicted in Fig. 8.

5
 Fig. 6: Examples of different mixer configurations

LO

Fig. 7: Two circuit configurations interchanging with the frequency of the LO where the switches repre-
sent the diodes.

LO
IF
RF

Fig. 8: Time-domain response of a mixer

6
 The output signal is always in the “linear regime”, provided that the mixer is not saturated with
respect to the RF input signal. Note, with respect to the LO signal the mixer has to be always in saturation
to insure the diodes operate almost as an ideal switch. The phase of the RF signal is conserved in the
output signal available at the IF output.

2.4 Amplifier
A linear amplifier, sometimes called “gain stage”, auguments the input signal by a factor which is usu-
ally indicated in decibels (dB). The ratio between the output and the input signals is called the transfer
function and its magnitude – the voltage gain G – is measured in dB and given as
VRFout VRFout
G[dB] = 20 · or = 20 · logG[lin]. (7)
VRFin VRFin
The circuit symbol of an amplifier is shown in Fig. 9 together with its S-matrix.

( )
0 0
S=
1 2 G 0

Fig. 9: Circuit symbol and S-matrix of an ideal amplifier

The bandwidth of an amplifier specifies the frequency range where it is usually operated, see
Fig. 10. This frequency range is defined by the −3 dB points1 of the magnitude response with respect
to its maximum or nominal transmission gain, dividing the magnitude transfer function of the amplifier
into a pass-band and a stop-band of equal transmitted power.
For an ideal amplifier the output signal would always be proportional to the input signal. However,
a real amplifier is non-linear, typically for larger signals the transfer characteristic deviates from its linear
properties, which is validated for small-signal amplification. When increasing the output power of an
amplifier, a point is reached where due to the non-linearities the small-signal gain is reduced by 1 dB
(Fig. 11). This output power level defines the so-called 1 dB compression point, which is an important
measure of the output power capability, thus the dynamic range for the amplifier.
The transfer characteristic of an amplifier can be described in commonly used terms of RF engi-
neering, i.e. the S-matrix, see Section 5. As implicitly contained in the S-matrix, both, amplitude and
phase information of any spectral component are preserved when passing through an ideal amplifier. For
a real amplifier the element G = S21 (transmission from port 1 to port 2) is not a constant, but a complex
function of frequency. Also the elements S11 and S22 are not zero.

Fig. 10: Definition of the bandwidth

1
The −3 dB points are the values left and right of a reference value, typically the local maximum of the amplifier transfer
function, and are 3 dB below that reference.

7
 Fig. 11: Example for the 1 dB compression point [4]

2.5 Interception points of non-linear devices

Important characteristics of non-linear devices are the interception points. Here, only a brief overview is
given, further information can be found in [4].
The most relevant interception points is the interception point of third order (IP3 point). Its im-
portance derives from its straightforward determination, plotting the input versus the output power on
a logarithmic scale (Fig. 11). The IP3 point is usually not measured directly, but is extrapolated from
the data, measured at much lower power levels in order to avoid overload or damage of the device under
test (DUT). Applying two signals (f1 , f2 > f1 ) of closely spaced frequencies ∆f simultaneously to
the DUT, the intermodulation products appear at +∆f above f2 and −∆f below f1 . This method is
called the third-order intermodulation (TOI). An example of an automatized TOI measurement is shown
in Fig. 12.
The transfer function of weakly non-linear devices can be approximated by a Taylor expansion.
Using n higher order terms and plotting them together with an ideal linear device on a logarithmic
log
scale results in two straight lines with different slopes (xn → n · log x). Their intersection point is
the intercept point of nth order. These points provide important information concerning the quality of
non-linear devices.
In this context, the aforementioned 1 dB compression point of an amplifier is the intercept point
of first order. For the method of measurements of the 1 dB compression point, see Section 7.4.
Similar characterization techniques can also be applied for mixers, which, with respect to the LO
signal, cannot be considered as weakly non-linear devices.

8
 Fig. 12: An example of automatized TOI measurement

2.6 The superheterodyne concept

The word superheterodyne is composed of three parts: super (Latin: over), ϵτ ϵρω (hetero, Greek: differ-
ent) and δυναµισ (dynamic, Greek: force), and can be translated as two forces superimposed2 . Differ-
ent abbreviations exist for the superheterodyne concept. In the USA it is often abbreviated by the simple
word “heterodyne”, and in Germany the shorter terms “super” or “superhet” are used.

RF amplifier Mixer IF amplifier Audio amplifier

BP

Bandpass filter Demodulator

Local oscillator (sometimes locked to a quartz crystal)

Fig. 13: Schematic drawing of a superheterodyne radio receiver

A “weak” incident (RF) signal is subjected to non-linear superposition (i.e. mixing or multiplica-
tion) with a “strong’‘ sine wave signal from a LO. At the mixer output sum and difference frequencies
of the RF and LO signals appear. The LO signal can be tuned such that this IF output signal is always
2
The direct translation (roughly) would be: another force becomes superimposed.

9
of same frequency, or stays within a very narrow frequency band. Therefore, a fixed-frequency bandpass
with excellent transfer characteristics can be used, which is cheaper and easier to realize than a variable
bandpass of the same performance. Also, gain-stages (amplifiers) operating at a lower IF frequency are
of better quality and/or are more affordable. A well-known application of this principle is any simple
radio receiver (Fig. 13).

3 Spectrum analyser
RF spectrum analyzers can be found in virtually every control room of a modern particle accelerator.
They are used for many aspects of beam diagnostics including Schottky signal acquisition and obser-
vation of RF signals. A spectrum analyzer is in principle very similar to a common superheterodyne
broadcast receiver, except with respect to the choice of functions, change of parameters, and in general
a more sophisticated, high quality design. It sweeps automatically through a specified frequency range,
which corresponds to an automatic turning of the tuning knob on a radio. The signal is then displayed
in the amplitude/frequency plane. Originally, these kind of measurement instruments were setup man-
ually and used a cathode ray tube (CRT) as display. Nowadays, with the availability of low-cost, pow-
erful digital electronics for control and signal processing, basically every instrument can be remotely
controlled. A microprocessor permits fast and reliable settings of the instrument, and an analog-digital-
converter (ADC) in connection with digital signal processing hardware performs the acquisition and
pre-processing of the measured signal values. The digital data processing enables extensive data treat-
ment for error correction, complex calibration routines and self tests, which are a great improvement
for RF signal measurements. However, the user of such sophisticated systems may not always be aware
of the basic analogue signal path and processing, before the signals are digitized and prepared for user
interaction. The basics of these analogue sections is discussed as follows.
In general, we distiguish two types of spectrum analyzers:
– the scalar spectrum analyzer (SA) and
– the vector spectrum analyzer (VSA).
The SA provides only information of the amplitude of the applied signal, while the VSA provides infor-
mation of the phase as well.

3.1 Scalar spectrum analyzer

A common oscilloscope displays a signal in the amplitude-vs.-time format (time domain). The SA
follows a different approach and displays the RF signal in the frequency domain.

A [V] A [dB]
0

t
– 40

t
2% AMPLITUDE MODULATION 2% AM IN FREQUENCY DOMAIN
Fig. 14: Example of amplitude modulation in time and frequency domains

One of the major advantages of the frequency-domain visualization lies in the higher sensitivity to
perturbations of periodic signals. For example, a 2% distortion of a sine-wave signal is already difficult

10
 TUNABLE AMPLITUDE
SIGNAL DISPLAY
BANDPASS DETECTOR

Fig. 15: A tunable bandpass as a simple spectrum analyser (SA)

MIXER
SIGNAL SWITCHABLE
LOW PASS IF FILTER
INPUT ATTENUATOR

TUNABLE
SAW TOOTH OSCILLATOR
IF
LO
AMPLIFIER
GENERATOR

VIDEO
VIDEO FILTER AMPLITUDE
DISPLAY
LOW PASS DETECTOR
AMPLIFIER

Fig. 16: Block diagram of a spectrum analyzer

to be observed on a the time domain display, but in the frequency domain on a logarithmic magnitude
scale the related “harmonics” (Fig. 14) are clearly visible (here −40 dB below the main spectral line). A
very faint amplitude modulation (AM) of 10−12 (power) on some sinusoidal signals would be completely
invisible on a time domain trace, but can be displayed as two side harmonics 120 dB below the carrier in
the frequency domain [5].
In the following we consider only “classical” SAs, based on a swept tuned band-pass filter analysis
(Fig. 15), or utilizing the heterodyne receiver principle (Fig. 16).
The simplest form of a swept frequency spectrum analyzer is based on a tunable bandpass. This
may be a classical lumped element LC circuit or a YIG filter (YIG = yttrium iron garnet) for frequen-
cies >1 GHz. The LC filter exhibits poor tuning, stability and resolution. YIG filters are used in the
microwave range (as preselectors) and for YIG oscillators. Their tuning range is about one decade, with
Q values exceeding 1000.
For superior performance, the superheterodyne principle is applied basically in all commercial
spectrum analyzers (Fig. 13). As already mentioned, the non-linear element (four-diode mixer or double-
balanced mixer) delivers mixing products, like

fsignal = fRF = fLO ± fIF . (8)

Assuming an input frequency range fRF from 0 to 1 GHz for the spectrum analyzer shown in Fig. 16 and
fLO ranging between 2 and 3 GHz, results in a frequency chart as shown in Fig. 17.
Obviously, for a wide range of input frequencies, while rejecting any image response, requires a
sufficiently high IF. A similar situation occurs for AM- and FM-broadcast receivers (AM-IF = 455 kHz,
FM-IF = 10.7 MHz). But, for a high IF (e.g. 2 GHz) a stable, narrowband IF filter is very challenging,

11
 5 (+)

3 fRF

2 (–)

2 fLO 3 GHz
Fig. 17: Frequency chart of the SA of Fig. 16, fIF = 2 GHz

therefore most SAs and high-quality receivers use more than a single IF. Certain SAs have four different
LOs, some fixed, some tunable. To perform a large tuning range, the first, and for fine tuning (e.g. 20 kHz
range), the third LO are variable.
Multiple mixing stages may also be necessary when downconverting to a lower IF (required when
using high-Q quartz filters) to ensure a good image response suppression of the mixers.
It can be demonstrated that the frequency of the nth LO must be higher than the (say) 80 dB
bandwidth (BW) of the (n − 1)th IF band-pass filter. A disadvantage of multiple mixing is the pos-
sible generation of intermodulation lines if amplitude levels in the conversion chain are not carefully
controlled.
The requirements of a modern SA with respect to frequency generation and mixing are

– high resolution,
– high stability (drift and phase noise),
– wide tuning range,
– no ambiguities

and, with respect to the amplitude response

– large dynamic range (>100 dB),

– calibrated, stable amplitude response,
– low internal distortions.

It is important to notice that the bandwidth ∆f of the IF band-pass filter is linked to sweep rate (or step
width and rate when using a synthesizer):
df
< (∆f )2 . (9)
dt
In other words, the signal frequency has to remain stable within ∆T = 1/∆f for a given IF bandwidth
∆f , which ensures steady-state conditions of the selected IF filter.
On many instruments the proper relation between ∆f and the optimum sweep rate is selected
automatically, but it can always be altered manually (setting of the resolution bandwidth).
Caution is advised when applying, but not necessarily displaying, two or more strong (> 10 dBm)
signals to the input. Third-order intermodulation products may appear (generated at the first mixer or
amplifier) and could lead to misinterpretation of the signals to be analyzed.

12
 Fig. 18: Block diagram of a vector spectrum analyser

Spectrum analyzers usually have a rather poor noise figure of 20–40 dB, as they often do not use
pre-amplifiers in front of the first mixer (dynamic range, linearity). But, with a good pre-amplifier, the
noise figure can be reduced to almost that of the pre-amplifier. This configuration permits amplifier
noise-figure measurements with a reasonable resolution of about 0.5 dB. The input of the amplifier to be
tested is connected to the hot and cold terminations, and the two corresponding traces on the SA display
are evaluated [6–10].

3.2 Vector spectrum and fast Fourier transform analyzer

The modern vector spectrum analyser (VSA) is essentially a combination of a two-channel digital os-
cilloscope and a fast Fourier transformation (FFT) based spectrum display. The incoming signal is
down-converted, band-pass (BP) filtered, and passed to an analog-to-digital converter (ADC) (general-
ized Nyquist for BP signals; fsample = 2 · BW). Fig. 18 shows a typical, simplified schematic of a modern
VSA.
The digitized signal is split into I (in-phase) and Q (quadrature, 90 degree offset) components with
respect to the phase of some reference oscillator. Without this reference, the term “vector” would be
meaningless for a spectral component.
One of the great advantages of a VSA, it easily allows to separate AM and FM components.
An example of vector spectrum analyzer display and performance is given in Figs. 19 and 20. Both
figures were obtained during measurements of the electron cloud in the CERN Super Proton Synchrotron
(SPS).

13
 Fig. 19: Single-sweep FFT display similar to a very slow scan on a swept spectrum analyser

Fig. 20: Spectrogram display containing about 200 traces as shown on the left-hand side in colour coding.
Time runs from top to bottom.

4 Noise basics
The concept of “noise” was originally studied for audible sound caused by statistical variations of the
air pressure with a wide flat spectrum (white noise). It is now also used for electrical signals, with
the noise “floor” determining the lower limit of the signal transmission. Typical noise sources are:
Brownian movement of charges (thermal noise), variations of the number of charges involved in the

14
conduction (flicker noise) and quantum effects (Schottky noise, shot noise). Thermal noise is only emitted
by structures with electromagnetic losses, which, by reciprocity, also absorb power. Pure reactances do
not emit noise (emissivity = 0).
Different categories of noise have been defined:

– white, which has a flat spectrum,

– pink, being low-pass filtered and
– blue, being high-pass filtered.

In addition to the spectral distribution, the amplitude density distribution is also required in order to
characterize a stochastic signal. For signals generated by superposition of many independent sources,
the amplitude density has a Gaussian distribution. The noise power density delivered to a load by a black
body is given by Planck’s formula:
NL ( )−1
= hf ehf /kT − 1 , (10)
∆f

where NL is the noise power delivered to the load, h = 6.625 · 10−34 J s the Planck constant and k =
1.38056 · 10−23 J/K Boltzmann’s constant.
Equation (10) indicates a constant noise power density up to about 120 GHz (at 290 K) with 1%
error. Beyond, the power density decays and there is no “ultraviolet catastrophe”, i.e. the total integrated
noise power is finite.
The radiated power density of a black body is given as

hf 3
Wr (f, T ) = [ ]. (11)
c2 ehf /kT − 1

For hf ≪ kT the Rayleigh–Jeans approximation of Eq. (10) holds:

NL = kT ∆f, (12)

where in this case NL is the power delivered to a matched load. The noise voltage v(t) of a resistor R
with no load is given as
v 2 (t) = 4kT R∆f (13)
and the short-circuit current i(t) by

kT ∆f
i2 (t) = 4 = 4kT G∆f, (14)
R

where v(t) and i(t) are stochastic signals, and G is 1/R. The linear averages v(t), i(t) vanishes, impor-
tant are the quadratic averages v 2 (t), i2 (t). The available power (which is independent of R) is given by
(see also Fig. 21)
v 2 (t)
= kT ∆f. (15)
4R
from which the spectral density function is defined as [6]

Wv (f ) = 4kT R,
Wi (f ) = 4kT G, (16)
∫ f2
2
v (t) = Wv (f )df.
f1

15
 Fig. 21: Equivalent circuit of a noisy resistor terminated by a noiseless load

R1 , T 1 ′
Wv1

R2 , T 2 R2 , T 2
Ri , W v

′
Wv2 ′
Wv3

Fig. 22: Noisy one-port with resistors of different temperatures [6, 11]

A noisy resistor may be composed of many elements (resistive network). Typically, it is made from a
carbon grain structure, which has a homogeneous temperature. But if we consider a network of resistors
with different temperatures, and hence with an inhomogeneous temperature distribution (Fig. 22), the
spectral density function becomes
∑
Wv = Wvj = 4kTn Ri , (17)
j

where Wvj are the individual noise sources (Fig. 23), Tn is the total noise temperature, Ri the total
input impedance, and βj are coefficients indicating the fractional part of the input power dissipated in
the resistor Rj . For simplicity it is assumed that all Wvj are uncorrelated.
The relative contribution (βj ) of a lossy element to the total noise temperature is equal to the
relative dissipated power multiplied by its temperature:
∑
Tn = β 1 T1 + β 2 T2 + β 3 T3 + · · · = β j Tj (18)
j

A good example is the noise temperature of a satellite receiver, which is nothing else than a directional
antenna. The noise temperature of free space amounts roughly to 3 K. The losses in the atmosphere,
which is an air layer of 10 to 20 km height, causes a noise temperature at the antenna output of about 10
to 50 K. This is well below our room temperature of 290 K.
So far, only pure resistors have been considered. Looking at complex impedances, it is evident,
losses occur only from dissipation in Re(Z). The available noise power is independent of the magnitude

16
 Wv1 Wv2 Wvi Wv

Ri Ri
Wu

Fig. 23: Equivalent sources for the circuit of Fig. 22

of Re(Z) with Re(Z) > 0. For Figs. 22 and 23, Eq. (17) still applies, except Ri is replaced by Re(Zi ).
However, in complex impedance networks the spectral power density Wv becomes frequency dependent
[11].
The rules mentioned above apply to passive structures. A forward-biased Schottky diode (external
power supply) has a noise temperature of about T0 /2 + 10%. A biased Schottky diode is not in thermody-
namic equilibrium and only half of the carriers contribute to the noise [6]. But, it represents a real 50 Ω
resistor when properly forward biased. For transistors, in particular field-effect transistors (FETs), the
physical mechanisms are somewhat more complicated. Noise temperatures of 50 K have been observed
for a FET at 290 K physical temperature.

4.1 Noise-figure measurements with the spectrum analyzer

Consider an ideal (noiseless) amplifier, terminated at its input (and output) with a load at 290 K with an
available power gain (Ga ). At the output we measure [7, 12]:

Pa = kT0 ∆f Ga . (19)

For T0 = 290 K (sometimes 300 K), we obtain kT0 = −174 dBm/Hz (−dBm = decibel below 1
mW). At the input we determine for a given signal Si a certain signal-to-noise ratio Si /Ni , and at the
output So /No , from what the noise factor F is defined as:
Si /Ni
F = (20)
So /No
and its logarithmic equivalent NF follows as:

NF = 10 log F (21)

An ideal amplifier has F = 1 or NF = 0 dB. The noise temperature of this amplifier is 0 K, and
signal and noise levels at the output are linearly increased by the gain. A real amplifier adds some noise,
which leads to a decrease in So /No due to the added noise Na :
N a + N i Ga Na + kT0 ∆f Ga
F = = . (22)
N i Ga kT0 ∆f Ga

For a linear system the minimum noise factor amounts to Fmin = 1 or NFmin = 0 dB, however,
for non-linear systems one may experience a noise factor F < 1.
Noise factor and noise temperature are related by
Na
Te = = T0 (F − 1) (23)
k∆f Ga

17
 + NOISE FREE
Pout
Power output (W) + Slope ≈ kGa ∆f

Noise free

Na
T0 = 290 K Thot = 12000 K

Source temperature Ts [K]

Fig. 24: Relation between source noise temperature Ts and output power Pout for an ideal (noise-free)
and a real amplifier [7, 12].

with Te being the equivalent temperature of a source impedance into a perfect, noise-free device that
would produce the same added noise Na [12].
The so-called Y -factor method is a popular way to measure the noise figure. It is based on a
switchable noise source with two calibrated values N1 and N2 for the noise temperature, e.g. Tc and Th ,
corresponding to “cold” and “hot”. Usually a dedicated noise diode is used as noise source, switched
between non-bias and bias operation to provide the two noise temperatures. The calibrated noise level is
defined as excess noise ratio (ENR):
( )
Th − Tc
ENR dB = 10 log (24)
T0
For most noise figure calculations the linear form is more useful:
ENRdB
ENR = 10 10 (25)

The noise source is connected to the amplifier or DUT to be analyzed, providing noise “on” (N2 )
and “off” (N1 ) conditions. The ratio of these noise powers is called the Y -factor:
N2
Y = (26)
N1
Y -factor and ENR can be used to determine the noise slope of the DUT, as illustrated in Fig. 24. The
calibrated ENR of the noise source represents a reference level for the input noise, which allows the
calculation of the internal (added) noise Na of the DUT:
( )
ENR
Na = kT0 ∆f G1 −1 (27)
Y −1

18
The SA, operating in automatized noise figure mode, controls the noise diode, i.e. switching between
“hot” (on) and “cold” (off) states, acquiring the DUT output signal, and computes – based on the cali-
brated ENR – the total system noise factor
ENR
Fsys = (28)
Y −1
which includes noise contributions from all parts of the system. In case the “cold” noise temperature
Tc ̸= T0 = 290 K, Eq. 28 becomes

ENR − Y (Tc /T0 − 1 )

Fsys = (29)
Y −1
For low ENR noise sources, Th < 10 Tc , an alternative equation holds:

ENR(Tc /T0 )
Fsys = (30)
Y −1

If Y is close to 1, i.e. Fsys ≫ ENR, the system noise factor “masks” the noise generated by the
noise source, making an accurate measurement difficult or impossible. Therefore the Y -factor method is
limited to noise figure measurements with NF ≈ 10 dB below the ENR of the noise source.
The literature explains a variety of other noise figure measurement methods [6,8–10,13], including
the “3 dB” method [12] for the measurement of high noise figure devices, where the Y -factor method is
limited.
The noise figure of a cascade of amplifiers is given as [6, 7, 11–13]
F2 − 1 F3 − 1
Ftotal = F1 + + + ··· . (31)
Ga1 Ga1 Ga2
As Eq. (31) shows, the first amplifier in a cascade has a dominant effect on the total (system) noise figure,
provided Ga1 is not too small and F2 not too large. In order to select the best amplifier from a number of
different units to be cascaded, the noise measure M
F −1
M= . (32)
1 − (1/Ga )

helps to select the optimal unit:

The amplifier with the smallest M should be selected as first unit in the cascade [12].

5 Introduction to network analysis and S-parameters

One of the most common measurement tasks in the field of RF engineering is the analysis of circuits and
electrical networks. Such networks can be a simple one-port (two-pole), containing only a few passive
components (resistors, inductors and capacitors) or they may be complex units, consisting of passive,
active and/or non-linear components with several input and output ports.
A vector network analyzer (VNA) is one of the most versatile and valuable pieces of measurement
equipment used in a RF laboratory or particle accelerator control room. The network analysis is per-
formed by exciting the device under test (DUT) with a well-defined input signal in terms of frequency
and amplitude, and recording the response of the network, for each frequency step as complex value of
the reflection and/or transmission coefficients. These are the coefficients of the scattering parameters
(S-Parameter), the properties to characterize a DUT at RF and microwave frequencies. The best com-
mercially available network analyzers can cover a frequency range of ten (and more) orders of magnitude
(from a few Hz to many GHz), with a resolution down to 0.1 Hz.

19
 (a) (b)

Fig. 25: Wave quantities of a one-port (with two poles) and impedance ZL : (a) incident (a1 ) and reflected
(b1 ) wave; (b) relation of a1 and b1 to V1 and I1 .

In the following sections, scalar and vector network analyzers are introduced and measurement
techniques for the determination of S-parameters of networks are discussed. S-parameters are basically
defined only for linear networks. In the real world, many DUTs are at least weakly non-linear (e.g.
mixers, or active elements such as amplifiers). For the analysis of these devices certain approximations
or extensions of the definitions are required [15].
Another interesting application is the determination of the beam transfer function (BTF), where
the DUT is a circulating particle beam in an accelerator.

5.1 One-port networks

In RF engineering, wave quantities are preferred in favor currents or voltages for the characterization of
RF circuits. We can distinguish between incident (a) and reflected waves (b). The incident wave trav-
els from a source to the DUT – the reflected wave travels in the opposite direction. This terminology
is preferred, because in RF engineering the linear geometrical dimensions of a circuit often are larger
than 10% of the corresponding free-space wavelength. Wave functions are defined in time and spacial
coordinates, and for this fact are preferred to voltages and currents, which typically are only defined in
time. This also requires the definition of a reference plane, i.e. the physical location in space to which the
measurement refers. Without this reference plane, e.g. the phase of the reflection coefficient would be
undefined, which would make vectorial measurements impossible. Of course, a mathematically correct
description of the DUT in terms of voltages and currents still holds, and also will return correct results,
but working with wave quantities turns out to be much more convenient in practice. Both network dis-
cription methods – if correctly applied – have no fundamental limitation, e.g. S-parameters can be used
at very low frequencies and voltage and current descriptions can also be used at very high frequencies.
Both methods are fully equivalent, for any frequency; the results are mutually convertible. This fact is
expressed by conversion rules, namely S-parameters can be converted into impedances and vice versa.
The interface of the DUT to the outside world is utilized by one or more pole pairs, which are
commonly referred as ports. A device with one pair of poles (as in Fig. 25a) is defined as one-port,
where one incident (a1 ) and one reflected (b1 ) wave can propagate simultaneously. The index of the
wave quantities represents the number of the port.
The wave quantities can be determined from the voltage and current at the port. They are related
to each other
V1 + I 1 Z 0 V1 − I 1 Z 0
a1 = √ , b1 = √ , (33)
2 Z0 2 Z0
where V1 and I1 represent the voltage and current respectively at the port as depicted in Fig. 25b. Z0 is
an arbitrary reference impedance (often, but not necessarily always, the characteristic impedance Z0 =
ZG = 50 Ω of the system).

20
 Fig. 26: All possible S-parameters of a two-port network

√
The wave quantities have the dimension of W (see [14]). This normalization is important for the
conservation of energy. The power traveling towards the DUT is calculated by Pinc = |a|2 , the reflected
power by |b|2 . It is important to note that this definition is mainly used in the USA – in European notation,
the incident power is usually calculated by Pinc = 0.5|a|2 . These conventions have no impact on the
calculation of S-parameters and only need to be considered when the absolute power is of interest.
The reflection coefficient Γ represents the ratio between the incident wave and the reflected wave
of a specific port. It is defined as
b1
Γ= . (34)
a1
By substitution with Eq. (33), we can find a relation between the complex (load) impedance ZL of
a one-port and its complex reflection coefficient Γ:
ZL − Z0
Γ= . (35)
ZL + Z0

5.2 Two-port networks

For electrical networks with two ports (e.g. attenuators, amplifiers) we find more quantities to be mea-
sured. Besides the reflection coefficients for each port, the transmission in forward and reverse directions
also needs to be characterized. We now require the definition of the scattering parameters (S-parameters)
for two ports. The idea is to describe how the incident energy on one port is scattered by the network
and exits through the other ports. All possible signal paths through a two-port are shown in Fig. 26. A
two-port has four complex, frequency-dependent scattering parameters:

b1 b1 b2 b2
S11 = , S12 = , S21 = , S22 = . (36)
a1 a2 a1 a2
Here S11 and S22 are equal to the reflection coefficients Γ of their respective ports – but only under the
condition that the corresponding other port is terminated in its characteristic impedance. S21 and S12
are the forward and reverse transmission coefficients, respectively. The first index of the S-parameter
defines at which port the outgoing wave is observed, the second index defines at which port the net-
work is excited. This leads to the counterintuitive appearing situation, that for forward transmission the
corresponding S-parameter is S21 , not S12 . The S-parameters are measured following exactly the same
definition. The internal source of the network analyzer excites an incident wave on port one, namely a1 .
Now b1 and b2 , the outgoing waves from the DUT, are measured, which allows the determination of S11
and S21 (provided that port one and port two are terminated with their characteristic impedances).
It is very important to always terminate all ports of the DUT with their respective characteristic
impedances. In many situations this is Z0 , but there are cases where the characteristic impedance is
different between port one and port two, e.g. a transformer with a turns ratio of two, leading to an
impedance transformation by a factor of four. In this case the characteristic impedance would be for port
one 50 Ω and for port two 12.5 Ω.

21
 The termination prevents unwanted reflections and ensures the DUT is only excited by a single
incident wave. For practical S-parameter measurements this implies that any port of the DUT needs
to be connected to a matched load corresponding to the characteristic impedance of this port. This
rule includes in particular the port connected to the VNA output port, or in other words, the generator
impedance has also to match the impedance of the DUT. For example, the analysis of a DUT with 25 Ω
characteristic impedance is not simply straightforward on a 50 Ω network analyzer, unless special care is
taken.. But, permitting a modern VNA, applying a special calibration procedure allows the modification
of the characteristic impedance of each VNA port to any value (within a reasonable range from > 5 Ω
to < 500 Ω), and in this way to adapt to the requirements of the DUT. However, the situation of the
termination of ports becomes more complicated for the characterization of beam elements, like beam
pickups, kickers, and accelerating structures, where strictly speaking the beam (waveguide) ports also
need to be terminated in their characteristic impedance. Often simple solutions can be applied, like
microwave absorbing foam, to avoid unwanted reflections from open beam ports.
The S-parameters are an intrinsic property of the DUT and not a function of the incident power
used for the measurement (condition of linearity). Obviously, the S-parameters measured shall be inde-
pendent of the instrumentation used to perform the measurement.
Once all n2 S-parameters for a given n-port network are measured, the properties of this network
can be described by a set of linear equations. For incident waves a1 and a2 of arbitrary phase and
magnitude on a two-port, the outgoing or scattered waves b1 and b2 can be determined

b1 = S11 a1 + S12 a2 , (37)

b2 = S21 a1 + S22 a2 .

These equations can be written in matrix format, for convenience:

⃗b = S ⃗a (38)
[ ] [ ][ ]
b1 S11 S12 a1
= . (39)
b2 S21 S22 a2

The S-matrix is a linear model of the DUT. Its diagonal elements represent the reflection coeffi-
cients of each port. The remaining elements characterize all possible signal transmission paths between
the ports. S-parameters are in general complex and a function of frequency. The set of linear equations
given by the S-matrix must be solved for a single frequency at a time. S-parameters are typically acquired
over a certain frequency range (span) for a number N of discrete, equidistant frequency steps. With N
data points, the system of equations has to be solved N times. A discussion of the general properties of
the S-matrix can be found in [14].

6 Scalar network analysis

A scalar network analyzer measures only the amplitude, i.e. the magnitude of a – reflected or transmitted
– signal, the phase is not available. Consequently, only the absolute value (the magnitude) of the complex
S-parameters can be obtained. Today scalar network analyzers are basically obsolete, however, some key
components and circuits are also found in VNAs, making this instrument a methodical way to introduce
the concept of network analysis.
A simple network analysis set-up, as it was used more than 50 years ago, is shown in Fig. 27.
The measurement is performed in two steps, in the first step (Fig. 27, left) without the DUT to measure
the power of the incident signal (V1 ). Then the DUT is inserted (Fig. 27, right), and V2 is measured.
Following the magnitude of the transmission coefficient is calculated:
V2
|S21 | ∝ . (40)
V1

22
 Fig. 27: A simple measurement set-up for the scalar transmission coefficient (|S21 |)

To obtain the results in decibels, a logarithmic amplifier was connected to the output of the detec-
tor. It has a logarithmic transfer function (Vout = log Vin ) and permits the display of a large dynamic
range on a dB scale. Furthermore, mathematical operations like multiplication or division, e.g. required
for normalization in Eq. (40), transforms simply into into an addition or subtraction, handled by opera-
tional amplifiers.
As detector any kind of device converting the input RF signal into a DC voltage is applicable,
assuming its transfer function is “reasonable”3 proportional to the RF power. There are basically three
possibilities to achieve this:
Rectifier A fast Schottky diode and a low-pass filter are used to convert the input RF signal to a DC volt-
age. Operating the diode in its square-law region (Pin < −10 dBm) results in an output voltage
proportional to the RF power; see Section 2.2.
Advantages: cheap, fast response (depending on fmax of the output filter).
Limitations: Commercially available RF power meters, based on Schottky diodes, can operate
from −60 dBm (limited by tangential sensitivity) up to about +30 dBm (damage level). The non-
linearity of the output signal versus input power is compensated by electronic means (look-up ta-
ble). Coaxial RF Schottky detectors are usually limited to maximum frequencies of approximately
100 GHz, essentially determined by the coaxial connector technology available. Usually an input
matching network is required to match the input impedance of the Schottky diode to Z0 = 50 Ω.
Thermal measurement Several types of detectors based on heating effects are available for the mea-
surement of RF power. In a bolometer (thermistor or barretter), the high temperature coefficient of
the thermal conductivity of certain metals or metal alloys is exploited. The temperature change ∆T
of dissipated heat of the RF input signal is measured utilizing a DC-based temperature measure-
ment, while applying a correction of the non-linearities. Barretters utilize the positive temperature
coefficient of metals like tungsten and platinum. Thermistors consist of a metal oxide with a
strong negative temperature coefficient. Another class of RF power meters based on heating is the
thermo-element, which takes advantage of the thermo-electrical coefficient of a junction between
two different metals. A well-known example is the Sb-Bi junction, which has a temperature co-
efficient of about 10−4 V/K, which is one of the highest values available for this kind of detector.
Even larger values can be achieved using semiconductor–metal junctions, where thermoelectric
coefficients of 250 µV/K have been achieved. For further details, see [16].
Mixer Multiplying two sinusoidal signals with different frequencies results in signals of sum and dif-
ference frequencies at the multiplier’s output; see Section 2.3. Technically this frequency mixing
principle allows to convert a range of high-frequency signals to a much lower intermediate fre-
quency (IF) band. Now the RF power measurement is performed in simpler ways at this IF.
3
With the term “reasonable” we point out the fact, that many detectors have a non-linear relation between input power and
output voltage.

23
 Fig. 28: Simplified circuit diagram of a typical automatic gain control

Fig. 29: Feedback loop of a typical automatic gain control (AGC)

6.1 Automatic Gain Control (AGC)

Often RF measurements are performed over a wide range of frequencies, requiring the signal strength,
i.e. the amplitude V0 of the source to be constant. This is usually achieved by an active feedback loop
(levelling), keeping V0 constant, independent of the operation frequency. Any feedback loop requires a
process variable which has to be detected and controlled to a well defined set point, here the output signal
level V1 . For the automatic gain control (AGC) loop in a NA e.g. a resistive power divider can be used to
provide this reference signal, while keeping inputs and outputs matched to Z0 = 50 Ω (Fig. 28). For this
example, the test signal arriving at the DUT is reduced by 6 dB due to the insertion loss of the resistive
power divider. However, the AGC feedback loop ensures the stimulus signal applied to the DUT has
always a constant, well defined power level over a wide frequency range.
For the characterization of linear DUTs, only the ratio V2 /V1 is of interest, which is independent
of the absolute value of V0 . In this case the S-parameter measurements do not require an AGC loop of
the RF generator, but in practice the gain control has many advantages, in particular for measurements
on weakly non-linear elements, such as amplifiers.

6.2 Directional couplers

Replacing the resistive power divider by a directional coupler reduces the insertion loss substantially,
the principle is outlined in Fig. 29. V1 is an attenuated replica – defined by the coupling factor – of the
forward-traveling wave, which is only used for as reference for the gain control. Typically, directional
couplers with a coupling coefficient of −20 dB are used for the purpose, they offer a transmission attenua-
tion in the main branch of less than 0.3 dB. In contrast to the resistive power splitter, the transmission-line
based directional coupler has a limited frequency range, and therefore other issues.
Modern network analyzers (both scalar and vectorial versions) measure the forward-transmission,
as well as the reflection coefficient of a DUT simultaneously, without the need to manually re-connect
DUT ports. Each port of the instrument is equipped with a dual directional coupler, providing simul-
taneously replicas of the incident and reflected waves from the DUT, see Fig. 30. These directional
couplers, in combination with some required switches and attenuators are commonly called test set. In
the early days, network analyzers consisted of separate building blocks, like S-parameter test set, fre-
quency generator, display and controller unit. All these elements had to be connected by many external
cables. Modern instruments have all those building blocks integrated in a single unit, including advanced

24
 Fig. 30: Dual directional coupler in a network analyzer

computer controls with digital data acquisition and post-processing.

Based on Fig. 30, the reflection and transmission coefficients are defined as
V3 V2
|S11 | ∝ , |S21 | ∝ . (41)
V1 V1

From the ratio of the reflected wave to the incident wave (S11 ), valuable quantities like standing
wave ratio (SWR), reflection coefficient, impedance, admittance as well as return loss of the DUT are
determined. From the ratio of the transmitted wave to the incident wave (S21 ), gain resp. insertion loss,
the transmission coefficient, the insertion phase, and group delay of the DUT can be characterized.

7 Vector measurements
A vector network analyzer (VNA) is able to measure the magnitude and phase of a complex S-parameter.
There are different hardware configurations which implement this kind of RF instrument, e.g. six-port
reflectometers, certain RF bridge methods, or superheterodyne RF network analyzers. Here only the
latter will be introduced.

7.1 The modern vector network analyzer

A modern VNA contains a RF generator which produces the signal stimulating the DUT. This signal
is usually generated by a synthesizer-type oscillator and is adjustable in very fine steps over a large
frequency range, in a programmable manner. Since all modern VNAs operate with analog and/or digital
downconverters (mixing), the generation of a tracking LO frequency is also necessary. This tracking LO
is typically generated by PLL circuits and represents essentially a second oscillator following the main
frequency with a specified frequency offset.
The observation (IF) band signal is typically processed digitally, allowing bandwidth settings over
a wide range, e.g. 1 Hz to 20 MHz and more. In all stages of the signal path the vectorial nature of
the signal is preserved, both phase and magnitude are processed, in the digital domain usually as I-Q
(in-phase – quadrature-phase) data, equivalent to real and imaginary parts. Details on the internal signal
processing of a VNA are found in [17, 18]. Note, similar to the spectrum analyzer, the sweep time and
resolution bandwidth cannot be adjusted independently. A modern four-port vector network analyzer is
shown in Fig. 31.
Although complete network analysis of any N -port can be performed with a two-port VNA, a
four-port unit is extremely convenient for many measurement tasks. It permits a quick analysis, e.g. of
a directional coupler or a three-port circulator without the need for swapping cables, it also introduces
virtual ports of balanced nature, and many other valuable features.

25
 Fig. 31: A modern four-port VNA

7.2 Time-domain transformation (synthetic pulse technique)

For any linear system, the frequency domain information (data) can be converted to the time domain
by an inverse (fast) Fourier transformation4 and vice versa, assuming the entire frequency vector data
(magnitude and phase, or real and imaginary) is present. This is the basis of the synthetic pulse technique,
available on many modern VNAs. It was commercially introduced by Hewlett-Packard in the 1980s for
network analyzer applications.
It renders the VNA even more versatile, allowing to display the impulse (Gaussian) and/or step
response of the DUT, and to perform time-domain reflectometry (TDR) measurements. Typical applica-
tions of this measurement techniques are:

1. Localizing and evaluating discontinuities (faults) in transmission lines.

2. Separating the scattering properties of sections of complicated RF networks by time-domain gat-
ing.
3. Echo cancellation (in multipath environments).
4. The synthetic pulse time-domain reflectometry can be very useful in trouble-shooting, e.g. of the
accelerator beam-pipe. By using waveguide modes it was successfully used to detect an obstacle
in the LHC beam-pipe.

The only constraint of the applicability of the synthetic pulse measurement technique, the DUT
has to be a linear and time-invariant (LTI) system.
A measurement example is shown in Fig. 32. A transmission line with a given length and some
perturbation is connected to a calibrated VNA. The real part of the Fourier-transformed reflection coeffi-
cient (S11 (ω)) is plotted versus time. The VNA permits the display of either, the synthetic step (Fig. 32a)
or the impulse response (Fig. 32b). The step is simply obtained by (numerical) integration of the impulse
response data.
The incident synthetic pulse is scattered from the discontinuity, but also from the open end of the
4
More precisely: by a discrete Fourier transformation (DFT). The fast Fourier transformation (FFT) is just an optimized
form of th eDFT, exploiting the symmetry of 2n data samples, thus saving computation time. However, both algorithms will
produce the same result for the same input data.

26
 (a) (b)

Fig. 32: Synthetic pulse measurement with a VNA: (a) step response; (b) impulse response.
The measured frequency data is converted by an inverse discrete Fourier transformation (iDFT) to the
time domain. Now the synthetic impulse response of a transmission-line, here a coaxial cable, is dis-
played over time. The reflections of the incident pulse on any irregularity or discontinuity, as well as the
end of the cable are clearly identified. By measuring the time delay between the the reference plane and
the location of the irregularity, or end of the cable (displayed as pulse or step in the reflection coefficient)
the electrical length of the cable can be calculated.

cable. The travel time for the pulse can be read on the horizontal axis on the time-domain display. In this
example we measure a delay of td = 22 ns until the open end of the cable becomes visible. This time
accounts for the impulse traveling towards the open end and back; thus, the factor 1/2 has to be taken
into account when calculating physical length l of the transmission-line:
c 1
l = √ · td . (42)
εr 2
In this example the relative dielectric constant of the insulation in the coaxial cable is ϵr = 2.3 (PTFE
Teflon), which returns a cable length of l = 2.2 m. The same method can be applied for obtaining the po-
sition of any irregularity or discontinuity (deformation, bad connector) along the cable. Nearly all VNAs
√
with time-domain option permit the designation of the velocity factor (1/ εr for a homogeneously filled
transmission line) and thus convert travel time or electrical length to physical distance on the display.
Note that the step response shown in Fig. 32a returns the local reflection factor versus time. Along
the cable it amounts to Γ = 0, except for the position of the irregularity, indicating a well-matched 50 Ω
transmission line. At the end we notice a positive step to Γ = 1, indicating an open circuit (see Table 2).
The reflected pulse in the impulse response trace (Fig. 32b), related to the open end of the cable
does not reach unit amplitude due to fact of cable attenuation of the transmission line used for this
example – a semi-rigid coaxial cable approximately 2 m length. The amplitude of this reflection from
the open end indicates the attenuation over twice the electrical length of the cable at the equivalent center
frequency (fmax = 3 GHz, fcentre = 1.5 GHz) of the measurement.
For practical applications of the synthetic pulse technique, certain basic properties of the discrete
Fourier transform should be kept in mind, they are summarized in Table 3. For example, a long cable
needs to be tested. This requires a long time window to ensure all multiple refections have decayed to
zero, which needs attention to ensure a sufficient narrow frequency sampling. The time interval ∆t is

27
 Table 2: Important values of the reflection coefficient

DUT ZL Γ
Open circuit ∞ +1
Short circuit 0 –1
Matched load Z0 0
Load Z0 /2 –1/3
Load 2Z0 1/3

Table 3: Important characteristics of the FFT

Time domain Frequency domain

Tmax (time span) ↔ ∆f (frequency resolution)
∆t (time resolution) ↔ fmax (frequency span)

related to 1/∆f , and this reciprocal relation may cause issues if settings are kept in “automatic” mode.
On the other hand, if a bad connector or cable damage needs to be located along a transmission-
line, a high resolution in time is required. Thus, the VNA has to measure over a wide frequency span
(fmax ). Obviously, we would like to often use both, a high frequency span and a close spacing of the
samples in the frequency domain, but there are practical limitations: namely, the number of data points
available. Usually in modern instruments the number of data points available amounts to 60000 and,
depending on the application, compromises have to be accepted.
Performing time-domain measurements with the vector network analyzer calls for two basic modes,
the “low-pass”, or the “band-pass” mode to be selected.

7.2.1 Low-pass mode

In low-pass mode the basic discrete Fourier transformation algorithm is applied. This returns certain
constraints on the frequency-domain measurement data of the DUT (Fig. 33a). The iDFT demands a
start frequency to always be 0 Hz (DC), and data is acquired in equidistant frequency steps between start
and stop frequency. Since most VNAs cannot measure at very low frequencies, the data points from
DC to the minimum operation frequency of the VNA are extrapolated mathematically. Data points for
negative frequencies are derived from the measured samples on the corresponding positive frequencies
by complex conjugation. Compared to the bandpass mode, this effectively doubles the number of data
points available for the calculation of the time trace. For this particular symmetry, the discrete Fourier
transformation returns a purely real-valued time trace. A practical time domain reflectometry (TDR)
measurement routine is setup as follows:

1. The DUT is connected, the port and type of measurement are selected (transmission or reflection).
2. The frequency range of interest and the number of data points are entered (this relates to the time
domain by Table 3)
3. After pushing the soft key, “set frequency low-pass”5 , the instrument choses the exact sampling
frequencies.
4. Once the sampling points are defined, the VNA has to be calibrated (open, short, load for reflection
measurements).

28
 (a) (b)

Fig. 33: Sampling of frequency points for the different operating modes: (a) low-pass mode; (b) bandpass
mode.

(a) (b)

Fig. 34: (a) Infinite frequency span. (b) Limited frequency span. The limited frequency span ∆f 6 of
the VNA leads to “distortions” of the time-domain synthetic pulse measurement. The ideal response is
convoluted with a sinc function, which characteristics depend on ∆f .

In the low-pass mode, the trace appearing on the screen for time domain reflectometry (TDR) or
time domain transmission (TDT) is basically equivalent to what a real-time TDR or sampling oscillo-
scope display; see Section 7.2.6.

7.2.2 Band-pass mode

In band-pass mode (Fig. 33b) the spectral lines (frequency-domain data points) no longer need to be
equidistant, and extrapolated down to DC, they just need to cover the frequency range of interest, e.g.
from fmin = 1.2 GHz to fmax = 1.5 GHz. The start and stop frequencies of the VNA can be chosen
arbitrarily, which returns a high degree of flexibility and is especially suited for the measurement of
devices having a limited frequency range (example: waveguide-mode reflectometry).
The bandpass mode is the equivalent to a narrowband TDR (and also time-domain transmission
TDT) using the synthetic pulse technique. It permits the display of the impulse response only, since no
extrapolated information on a DC component is available. The measurement clearly identifies position
and size of perturbations along a transmission line, including waveguides. Their characterization in
terms of capacitive, inductive or resistive properties is possible, but not straightforward [19]. Details
on the general properties and mathematical backgrounds of the low-pass and bandpass modes are found
in [18, 20].

7.2.3 Windowing
As the VNA always samples a limited frequency spectrum, starting at fmin and stopping at fmax , the
acquired spectrum is clipped by a rectangular envelope. Performing the iDFT, rectangular windowing
artifacts show up in the time-domain data, as compared in Fig. 34.
5
This soft key may appear with slightly different naming, depending on the definitions of the manufacturer.
6
not to be confused with the previous definition of ∆f for the equidistant frequency samples

29
 Fig. 35: Typical window functions to suppress strong sidelobes

An infinite spectrum of constant density (shown in Fig. 34a) leads to a Dirac-pulse function in
the time domain. The Dirac pulse contains by definition all frequency components of equal power. In
Fig. 34b, the spectrum is limited, for example, by the maximum operation frequency of the VNA, or
by some user settings. This can be expressed by multiplication of the ideal spectrum with a rectangular
function. The iDFT of a rectangular function of width ∆f leads to a sinc function (sometimes denoted
as si function) in the time domain. This relation is shown in Eq. (43) and graphically in Fig. 34.

Frequency domain ⇐⇒ Time domain

( )
f sin (∆f πt) (43)
rect ⇐⇒ = ∆f · sinc (∆f πt) .
∆f πt

To mitigate the effect of rectangular clipping of the spectrum in the time domain result, various weighting
functions are available. They smoothly filter (reduce) the amplitude of the spectrum around fmin and
fmax in band-pass and low-pass mode. This helps to reduce the strong sidelobes (ringing) in the time
domain. However, the price to be paid is a reduced pass-band, thus limiting the time resolution and the
ability to distinguish between two closely spaced impulses. The user has to select a reasonable trade off
between the window weighting functions, depending on the requirements of the particular measurement.
The effect of some window functions on main and sidelobes is shown in the frequency domain(!) on a
logarithmic scale in Fig. 35.

7.2.4 Gating
The gating option of the VNA allows to eliminate or select parts of the time-domain signal, provided
they are reasonably well separated in the time-domain trace.
For example, the already mentioned cable, connecting to the VNA port, is assumed to have an
internal irregularity at a certain position. By suitable selection of a time-domain gate (highlighted in

30
Fig. 36: Only the signal in a certain time window is of interest. After selection, the FFT of this window
will be calculated. Here the real values of the synthetic impulse response are shown on a linear scale.

Fig. 36 from t ≈ 18 ns to t ≈ 26 ns), the desired portion of the time domain trace (here, the total
reflection at the open cable end) can be separated from the the rest of the trace (set to zero). This allows
an analysis, e.g. by transformation back to the frequency domain, of the interesting part of the circuit
without influence of multiple reflections and perturbations from discontinuities, etc. (de-embedding).
For transmission measurements, usually the first arriving pulse in the time domain is selected, thus
suppressing the effect of all following reflections and related signals. For reflection measurements, the
first, but also following pulse response in the time-domain trace may be selected.
The implemented time-domain gating function is not a “brick wall”, but a soft switch applying a
weighting function similar to the iDFT window function. As it is a non-linear operation, it may generate
additional frequency components which were not present in the original signal. As general practical
guide line, the gate should not cut into a signal trace different from zero.

7.2.5 Examples of synthetic pulse time-domain measurements

A collection of measurement examples of simple DUTs are shown in Fig. 37. For all cases depicted, the
VNA is set up in step response operation. The traces from top to bottom show:

1. Matched load (Z = ZC ). As Γ is equal to zero, the response is zero everywhere.

2. Moderate (resistive) mismatch (Z = 2ZC , e.g. 100 Ω in a 50 Ω system). During the first 200 ps
the trace displays the well impedance-matched cable, following the reflection coefficient jumps to
a positive, constand value due to the impedance mismatch.
3. Capacitor. The TDR displays the capacitive load for a moment as a short circuit, and resumes with
an exponential function, as the capacitor is charged. The final state is equivalent to an open circuit,
as expected.
4. Inductor. In the TDR the inductive load appears at t = 200 ps as an open circuit, followed by
an exponential decay function. The steady state results in a short circuit, as the inductor is fully

31
Fig. 37: Examples of an arbitrary impedance, measured in TDR

32
 conducting.

7.2.6 Comparison to true time-domain measurements

There is a wide range of applications for the discussed synthetic pulse time-domain technique. A VNA
in time-domain low-pass step mode has a very similar range of applications as a TDR sampling oscillo-
scope. However, the synthetic pulse method is limited to strictly linear systems, therefore the analysis
of transient or non-linear systems, e.g. settling response of a microwave oscillator after power up would
not give very meaningful results. In other words, for highly non-linear and time-varying DUTs true
time-domain measurements, based on pulse generators and oscilloscopes are still indispensable, e.g. an
air traffic radar system, where we have linear but time-varying conditions.
The dynamic range of a typical sampling oscilloscope is limited to about 60 to 80 dB with a
maximum input signal of 1 V and a noise floor around 0.1 to 1 mV (typical broadband oscilloscope).
The dynamic range of the VNA is > 100 dB, allowing similar maximum input levels of approximately
+10 dBm (some VNAs allow +20 dBm). Both instruments are using basically the same kind of detector,
either a balanced mixer (four diodes) or a sampling head (two, four or six diodes), but the essential
difference lies in the noise floor and the average signal power arriving at the receiver input. In case of
the VNA the measurement is based on a continuous-wave (CW) signal with bandwidth of a few Hz, and
thus can obtain with appropriate filtering a very good signal-to-noise ratio7 .
A traditional sampling oscilloscope acquires the data during a short time with a rather low repeti-
tion rate (typically around 100 kHz up to a few MHz), with all the thermal noise power spread over the
entire frequency range (typically 20–50 GHz bandwidth). With this low average signal power (around a
microwatt) the signal spectral density is orders of magnitude lower compared to the VNA measurement
procedure (it acquires signals continuously), which explains the large difference in dynamic range (even
without gain switching).
A more detailed discussion about time-domain reflectometry with vector network analysers can
be found in [20].

7.3 Calibration methods

The hardware of even an “ultra-modern” VNA is not perfect, e.g. the internal source is not perfectly
impedance matched to 50 Ω (over the entire frequency range), its internal directional couplers have a
finite directivity, since there exists no ideal (infinite) directivity in practice, and finally the coaxial cables
between VNA and DUT ports have frequency-dependent attenuation (dispersion) effects.
This calls for a calibration to compensate all these unwanted effects, to guarantee a precise, instru-
ment independent analysis of the DUT. There are several calibration procedures to eliminate some, or all
of the mentioned deficiencies. The easiest is called the “response calibration”, typically applied for trans-
mission, rarely for reflection measurements. It basically is a S21 (or S12 ) transmission measurement of a
quasi “zero length” ideal transmission-line, by connecting the two cable ends of the two-port VNA with
each other. For the given VNA setting, i.e. start / stop frequency, # of freq. points, resolution bandwidth,
power level, etc., magnitude and phase are acquired and stored as S21reference in the non-volatile memory
for each frequency point. Now, a DUT can be connected between the cable ports, with the connectors
serving as reference planes of the calibrated system (VNA plus cables). In calibrated mode the VNA
performs:
S21DUTmeas.
S21DUTcal. = , (44)
S21reference
However, this simple calibration procedure eliminates essentially the frequency-dependent losses
and phase-transfer functions of the test cables only. But, the mismatch between cable and generator, and
7
Remember the thermal noise is proportional to measurement bandwidth. Its density at room temperature is –174 dBm/Hz.

33
Fig. 38: Error model of a VNA. The parameters exx of the error network are determined by the calibration
procedure and used to determine the true (corrected) result (ΓDUT ) based on the measured result (ΓM ).

Table 4: Interpretation of VNA error terms

Error term Interpretation

e10 Reflection tracking
e00 Directivity
e11 Test-port match

the impact of the finite directivity are still present. A more sophisticated, and widely popular calibration
technique for the reflection measurements needs to be performed: the open, short and match technique.
This technique covers the three independent error sources mentioned above: finite directivity, generator
mismatch and the transfer function of the cables.
The VNA applies an internal error model, shown in Fig. 38. The measured raw data acquired
by the instrument (ΓM ) is distorted by certain systematic errors. These errors are modeled via four
parameters: e10 , e00 , e01 , e11 , based on the error network model of Fig. 38. enn are in general complex
and frequency dependent parameters, furthermore e10 = e01 . The error parameters are extracted and
stored when performing a suitable calibration method, i.e. open, short, match, such that the true value
of the DUT (ΓDUT ) is calculated and presented accordingly. In simple terms, we need to carry out
three independent measurements for each frequency point, to solve three coupled equations with three
complex unknowns. These error terms represent the above-mentioned effects as listed in Table 4.
The unknowns of the error network are determined applying a calibration measurement with three
different, but known, calibration DUTs. These calibration DUTs do not need to be perfect, only the
electromagnetic properties need to be known with great precision. The tabulated complex, frequency-
dependent S-parameters of the calibration standards are provided by the manufacturer of the calibration
hardware (they are often referred as calibration kit), and are stored in the VNA memory as calibration
kit reference data. Usually the calibration DUTs represent an open circuit, a short circuit and a matched
load (termination), enabling the VNA to determine the frequency-dependent error model. This is altered
if different test cables are used, or if the VNA settings are modified, and would require a re-calibration
under those circumstances. Now the VNA continuously applies the error correction during the DUT
measurement, and the reference plane is “moved” to the end of the test cables. Only the DUT networks
“behind” the reference plane are taken into account for the measurement.
The impact of the VNA calibration is demonstrated in Fig. 39, which presents a S11 measurement
of a high-quality 50 Ω termination, with and without VNA calibration. For an ideal termination, no
reflection should be present, i.e. S11 = 0 ≡ −∞ dB. In this example the calibration of the VNA
improves the measurement quality by 20 dB! In case of a short (total reflection, S11 = 1 ≡ 0 dB), a

34
Fig. 39: S11 measurement of a 50 Ω termination with and without calibration. The calibration provides
20 dB improvement for this frequency range.

(a) (b)

Fig. 40: Typical calibration kits for a VNA: (a) manual (open, short, match); (b) electronic

non-calibrated S11 response typically displays a residual with values of a fraction of a dB, up to a few
dB below the 0 dB line (same for an open); after calibration these error reduces to a few millidecibels.
So far, we have covered the “response calibration” and the “complete one-port calibration”. To
perform completely error-corrected transmission measurements, the “full two-port calibration” proce-
dure has to be applied. Therefore, the error model is expanded to include the errors from the receiving
port, requiring a calibration of each port based on the just discussed “complete one-port calibration”
method. Also, for transmission, we need two standards, i.e. the “response calibration” and the “iso-
lation calibration”, however, latter often may be omitted. In summary, the “full two-port calibration”
consists out of a “complete one-port calibration” procedure for each port, which requires open, short and
match standards, plus the “response calibration” and eventually the “isolation calibration”. In total eight
calibration measurements have to be performed to bring the VNA into the desired CAL status.
For measurements on devices with standard coaxial connectors, e.g. SMA or N-type, calibra-
tion standards such as a termination, an open and a short circuit are available (shown in Fig. 40a). As
mentioned, to successfully perform the calibration procedure for the reflection coefficient, the tabulated

35
values, representing the electromagnetic properties of the calibration standards, has to be present in the
VNA. Obviously, the tabulated parameters of the calibration kit does not have an infinite frequency reso-
lution. The instrument applies an interpolation procedure if the selected frequency points are not exactly
at the tabulated values of the calibration kit.
The calibration technique described so far is a well established industry standard for RF and mi-
crowave VNA measurements. However, it has a substantial disadvantage for the user: it is tedious and
time consuming, in particular if a calibration of a multiport VNA is required.
Already for the full two-port calibration requires eight calibration measurements to satisfy the eight-
term error model. The manual procedure of connection and de-connection of the calibration standards is
time consuming, boring, and prone to errors. The situation becomes even worse when performing a full
four-port calibration (32 connections and de-connections of standards). For this reason, the electronic
calibration kit method is available and now very popular. For this procedure, each port is connected via
the measurement cable to the electronic calibration box (shown in Fig. 40b), which holds the different
calibration standards, and switches them automatically controlled by the VNA. This method enables to
perform a full four-port calibration in less than a minute. Again, like for the manual calibration method,
the standards do not need to be perfect, but well known, reproducible (switching) and stable. More
details are found in [17, 18].

7.4 1 dB compression point measurement

A single tone sine-wave source is connected to the input of an amplifier and its amplitude level is gradu-
ally increased versus time. Monitoring the output of this amplifier, we notice a proportional dependence
between input and output powers for small signal levels. This proportionality is referred as the linear
gain factor. For higher input signal levels, this relationship does not hold any more, since the amplifier is
not a perfectly linear system, and suffers from “saturation” effects. A fraction of the output power will
appear at other frequencies, which are higher order harmonics of the input signal. Typically the second
and third harmonics are dominant, and the related signal distortion is referred as harmonic distortion.
In parallel, we observe a compression of the gain for the fundamental signal. The actual gain falls off
below the small-signal gain response (Fig. 41). If this deviation amounts to 1 dB, we have reached the
“1 dB compression point”. Typically the industry refers to the output power, when specifying the 1 dB
compression point for their RF products.

Fig. 41: Definition of the 1 dB compression point for an amplifier: input vs. output power at the point
where the power level falls below 1 dB from its (linearly) predicted value.

36
 movable electric field probe

from
generator
DUT

Vmax

Vmin

Fig. 42: Schematic view of a measurement set-up used to determine the reflection coefficient as well as
the voltage standing wave ratio of a device under test (DUT) [21].

The 1 dB compression point is an important figure of merit, used to characterize the linearity of
a RF system, in particular the performance of small-signal and power amplifiers. It can be comfortably
measured with most VNAs in CW mode, i.e. choosing a single frequency and performing a power sweep.
In power sweep mode, the instrument displays a trace similar as shown in Fig. 41.

8 Introduction to the Smith chart

Even with today’s availability of computer-aided simulation and circuit simulation software suites, the
Smith chart is still a very valuable and important tool that facilitates an interpretation of the (half) complex
impedance plane with respect to the S-parameters, and the related calculations and measurements. This
section gives a brief overview of the concept, and more importantly, of how to use the chart. Its definition,
as well as an introduction of how to navigate on the chart are illustrated. Some typical examples illustrate
the broad range of applications of the Smith chart.

8.1 Voltage standing wave ratio (VSWR)

With modern RF measurement equipment available today it is rather easy to precisely measure the re-
flection factor Γ, even for complicated networks. In the “good old days” though, this was performed by
measuring the electrical field strength8 along a slotted coaxial line, which has a longitudinal slit to allow
a small field probe to be slided to any location along the line (Fig. 42). This electric field probe, protrud-
ing into the field region of the coaxial line near the outer conductor, picked up an E-field signal, which
was displayed on a microvoltmeter after rectification via a microwave diode. While moving the probe,
field maxima and minima, as well as their position and spacing where recorded. From this information
the reflection factor Γ and the voltage standing wave ratio (VSWR or SWR) were determined:

– Γ is defined as the ratio of the electrical field strength E of the reflected wave versus the forward-
traveling wave:
E of reflected wave
Γ= . (45)
E of forward-traveling wave
8
The electrical field strength was used, since its measurement was considerably easier than that of the magnetic field.

37
 X = Im (Z) Im (Γ)

R = Re (Z) Re (Γ)

Fig. 43: Illustration of the Moebius transformation from the complex impedance plane to the Γ plane,
commonly known as Smith chart.

– The VSWR is defined as the ratio of maximum to minimum measured voltages:

Vmax 1 + |Γ|
VSWR = = . (46)
Vmin 1 − |Γ|

Although today these measurements are far easier to conduct, the definitions of the aforementioned quan-
tities are still valid. On top, their importance has not diminished in the field of microwave engineering,
both reflection coefficient as well as VSWR are still a vital part of the everyday life of a microwave
engineer performing simulations or measurements.

8.2 Definition of the Smith chart

The Smith chart [22] provides a graphical representation of Γ that permits the determination of quantities
like the VSWR, or the impedance of a device under test (DUT). It uses the bilinear Moebius transforma-
tion, projecting the complex impedance plane on the complex Γ plane:
Z − Z0
Γ= with Z = R + j X. (47)
Z + Z0
As shown in Fig. 43, the half–plane with positive real part of impedance Z is mapped to the interior of
the unit circle of the Γ plane.

8.2.1 Properties of the transformation

In general, this transformation has two main properties:

– generalized circles are transformed to generalized circles (note that a straight line is nothing else
than a circle with infinite radius and is therefore mapped as circle to the Smith chart);
– angles are preserved locally.

Figure 44 illustrates how certain basic shapes transform between impedance and Γ planes.

38
X = Im (Z) Im (Γ)

R = Re (Z) Re (Γ)

Fig. 44: Illustration of the transformation of basic shapes from the Z to the Γ plane.

0.12 0.13 0.14

0.11 0.1
0
0.1 0.38 0.37 0.36 5
9 0.39 0.3 0.1
0.0 0.4
0
100
90 80 5 6
0.3 0.
. 08 0.41 110 0.9 1 1.2
70 4 17
0 0.8 0.
42 1.4 3
0. 120 0.7 60 3 0.
07 1.6 18
0. 3 .6 0.
4 0 1. 32
0. 0 8 50

0.
06

0.2
13 0.
5 2

19
0.

0.
44

31 40
14 0.

0.
05

20
0.4
0.
0.

0.
45

30
0.

0.
04

0.6
21
3

0.
6
0.

0
0.
.4

30
15

29
160 7 0

8
0.
4

0.2
3
0.0

0.2
0.4

1
0.2

8
20

0.23
0.02

0.27
0.48

0.1
0.1

10
170

10

0.24
0.01

0.26
0.49

05

20 50 -50-20 -10
0.

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9
1.0

1.2

1.4

1.6

1.8
2.0

3.0

4.0

5.0

0.25
0.25
0.00
0.00

10

20

50
180
0

0
-0.05

0.24
01

0.26
0.49

-170
0.

-1
.1

0
- 0

0.23
2

0
0.48
0

.
.

2
.
0

7
0

-20
6

0.2

-5
-1

0
03

1
0.2
7

2
.
4

2
0
0.

8
4

0.
8
0

0.
04

-
5 -15

.3

30
21

0
46
.

.
0

0.6
2
-
.

-3

9
0

0.
0

2
0
0.

.4

-4 0.

0.
5

0 14

0.4
4

-0

0 19

3
.

0
0

6
0.

0.
4

.5 -5 -2
4

30
31

-0 0.2
0.

-1 7 0. 0 -1
.8
0 18
0. -0
.6 0. -1
.6
20 43 32 0. -60
-0.
7 -1 .08 0. 17 -1.
4
0
-0.8 -1.2 0.
-11
0 9 42 -0.9 -1 33 0.1 -70
0.0 0. 6
0 -100 -80 0.3 0.1
0.1 0. 41 -90
0.14 4 5
0 0.11 0.13 0.12 0.3
0.4 0.36
5
0.39 0.38 0.37

Fig. 45: Example of a typical Smith chart

39
 Γ = − YY −Y
+Y0 with Y = G + j B
0

B = Im (Y ) Im (Γ)

G = Re (Y ) Re (Γ)

Fig. 46: Mapping of the admittance plane into the Γ plane

8.2.2 Normalization
The Smith chart is usually normalized to a reference impedance Z0 (= real):
Z
z= . (48)
Z0
This simplifies the transformation:
z−1 1+Γ
Γ= ⇔ z= . (49)
z+1 1−Γ
Although Z0 = 50 Ω is the most common reference impedance (typical characteristic impedance of
coaxial cables) and many applications use this normalization, any other real, positive value is valid.
Therefore, it is crucial to check the normalization assumed, before using any chart.
Being unfamiliar, the Smith charts appears confusing at a first look, with a fine grid from the
Z-plane mapped to a dense grid of many circles on the chart (Fig. 45).

8.2.3 Admittance plane

The Moebius transformation which generates the Smith chart also provides a mapping of the complex
admittance plane (Y = 1/Z, or normalized y = 1/z) into the same chart:
y−1 Y − Y0 1/Z − 1/Z0 Z − Z0 z−1
Γ=− =− =− = = . (50)
y+1 Y + Y0 1/Z + 1/Z0 Z + Z0 z+1
Using this transformation results in the same chart, but mirrored at the center of the Smith chart (Fig.
46). Often both mappings, the admittance and the impedance plane are combined into one chart, which
then looks even more overwhelming. For reasons of simplicity all illustrations in this article use only the
mapping from the impedance to the Γ plane.

8.3 Navigation in the Smith chart

The representation of circuit elements in the Smith chart is discussed in this section, starting with some
important points inside the chart. The following examples of circuit elements illustrate their representa-
tion in the chart.

40
 Im (Γ)

short circuit open circuit

Re (Γ)

matched load

Fig. 47: Important points in the Smith chart

8.3.1 Important points

There are three important points in the chart:

1. Open circuit with Γ = 1, z → ∞.

2. Short circuit with Γ = −1, z = 0.
3. Matched load with Γ = 0, z = 1.

They all are located along the real axis at the beginning and the end, which are also on the outer circle
(imaginary axis), and at the center of the Smith chart (Fig. 47). The upper half of the chart is “inductive”,
since it corresponds to the positive imaginary part of the impedance. The lower half is “capacitive”, as it
is corresponding to the negative imaginary part of the impedance.
Concentric circles around the center represent constant reflection factors (Fig. 48). Their radius
is directly proportional to the magnitude of Γ; therefore, a radius of 0.5 corresponds to reflection of 3
dB (half of the signal is reflected), whereas the outermost circle (radius = 1) represents total reflection.
Evidently, matching problems are clearly visualized in the Smith chart, since a mismatch will lead to a
reflection coefficient larger than 0, see Eq. (51).
1( 2 ) |a|2 ( )
Power into the load = forward power - reflected power: P = |a| − |b|2 = 1 − |Γ|2 . (51)
2 2
In Eq. (51) the European notation is used9 : power = |a|2 /2. Furthermore it should be noted, (1 − |Γ|2 )
corresponds to the losses due to the impedance mismatch.
Even though here we limit to the mapping of the impedance plane to the Γ plane, The admittance
is simple to determine, since
( )
1 1/z − 1 1−z z−1 1
Γ( ) = = = or Γ( ) = −Γ(z). (52)
z 1/z + 1 1+z z+1 z
In the Smith chart this fact is visualized as a 180◦ rotation of the vector of a given impedance (Fig. 49).
9
The commonly used notation in the USA: power = |a|2 . These conventions have no impact on the S-parameters, but they
are relevant for absolute power calculations. Since this is rarely used in context with Smith chart gymnastics, the actual power
definition used is not critical.

41
 |Γ| = 1
0.12 0.13 0.14
0.11 0.1
0 5
0.1 0.38 0.37 0.36
9 0.39 0.3 0.1

|Γ| = 0.75
0.0 0.4
0
100
90 80 5 6
1 0.3 0.
08 70
0. 0.4 110 0.9 1 1.2 4 17
0.8 0

|Γ| = 0.5
1.4
42
0. 120 0.7 60 .33 0.
07 1.6 18
0. 6 0.
43 0. 1. 3
0. 0 8 50 2

|Γ| = 0.25

0.
06
0.2
13 0.
5 2

19
0.

0.
0 44

31 40
14 0.
|Γ| = 0

0.
05

20
0.4

0.
0.

0.
45

30
0.

0.
04
0.6

21
3

0.
6
0.

0
0.
.4

30
15

29
160 7 0
0.8

0.2
3
0.0

0.2
0.4

2
1

.2

8
20
5 0

0.23
0.02

0.27
0.48

1 0.1

10
170

10
0 0.05 0.

0.24
0.01

0.26
0.49

20 50 -50-20 -10
0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9
1.0

1.2

1.4

1.6

1.8
2.0

3.0

4.0

5.0

0.25
0.25
0.00
0.00

10

20

50
180

0
-0.05

0.24
0.01

0.26
0.49

-170

-10
5 -0.1

0.23
0.02

0.27
0.48

-0.1
0.0 0

-2
-16

-0.2

-5

0.2
3

0
1

0.2
7
0.4

2
-4

8
0.8
50

0
04

-3
.3

.2

0.
46

05 -1

0
0.

-0

1
0.6

29
0.

-3

0.
20
0
.

.4

-4 0.

0.
45

06 4
0

0.4
0. -1

-0

0 19

30
0.

0.
44

.5 -5 -2
30

31
-0
.

0.2 0. 0
0

-1
-1 7 18 .8
0
-0
.6 0. 0. -1
.6
20 43 32 0. -60
-0.7 -1 .08 0. 17 -1.4
0
0 2 -0.8 -0.9 -1
-1.2 0.
33 0.1 -70
1 9 4
-1 0.0 0. 6
0 -100 -80 0.3 0.1
0.1 1 -90 5
0.4 0.11 0.14 4
0 0.13 0.12 0.3
0.4 5
0.39 0.36
0.37 0.38

Fig. 48: Illustration of circles representing a constant reflection factor

0.12 0.13 0.14

0.11 0.1
0 5
0.1 0.38 0.37 0.36
9 0.39 0.3 0.1
0.0 0.4
0
100
90 80 5 6
1 0.3 0.
08 70
0. 0.4 110 0.9 1 1.2 4 17
0.8 1.4 0
42
0. 120 0.7 60 .33 0.
07 1.6 18
0. 6 0.
43 0. 1. 3
0. 0 8 50 2
0.
06

0.2
13 0.
5 2
19
0.

0.
0 44

31 40
14 0.

0.
05

0.4 20
0.
0.

0.
45

30

Impedance z
0.

0.
04

0.6
21
3

0.
6
0.

0
0.
.4

30
15

29
160 7 0

0.8
4

Reflection Γ
0.2
3
0.0

0.2
0.4

1
.2

8
20
5 0

0.23
0.02

0.27
0.48

1 0.1

10
170

10
0 0.05 0.

0.24
0.01

0.26
0.49

20 50 -50-20 -10
0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9
1.0

1.2

1.4

1.6

1.8
2.0

3.0

4.0

5.0

0.25
0.25
0.00
0.00

10

20

50
180

0
-0.05

0.24
0.01

0.26
0.49

-170

-10
5 -0.1

0.23
0.02

0.27
0.48

-0.1
0.0 0

-2
-16

-0.2

-5

0.2
3

1
0.2
7
0.4

Reflection -Γ
2
-4

0.8
50

0
04

-3
.3

.2

0.
46

05 -1

0
0.

-0

0.6
29
0.

-3

Admittance y = z1
0.
20
0
.

.4

-4 0.

0.
45

06 4
0

0.4
0. -1

-0

0 19

30
0.

0.
44

.5 -5 -2
30
31

-0
.

0.2 0. 0
0

-1
-1 7 18 .8
0
-0
.6 0. 0. -1
.6
20 8 43 32 0. -60
-0.7 - 1 0 0. 17 -1.4
0.
-0.8 -1.2 0.
-11
0 9 42 -0.9 -1 33 0.1 -70
0.0 0. 6
0 -100 -80 0.3 0.1
0.1 . 4 1 -90 4 5
0 0.11 0.14
0 0.13 0.12 0.3
0.4 5
0.39 0.36
0.37 0.38

Fig. 49: Conversion of an impedance to the corresponding admittance in the Smith chart

8.3.2 Impedance of simple, passive lumped element circuits

Consider a simple passive circuit: a lumped, reactive element (inductance L, or capacitance C) of ar-
bitrary value connected in series to an resistance R. The corresponding signature of this circuit in the
Smith chart, varying the inductance resp. capacitance, is a circle. For a given type of impedance, the trace
of this circle follows a clockwise (inductance), or anticlockwise (capacitance) movement (Fig. 50). If a
lumped, reactive element is connected in parallel to R, the pattern is basically the same, but rotated by
180◦ (Fig. 51). It is equivalent to the discussed admittance mapping. Summarizing both cases, results in
a simple rule for the navigation in the Smith chart:

Reactive elements connected in series follow the trajectory of a circle in the impedance plane. Induc-
tances move clockwise, capacitances move anticlockwise when increasing their value. Reactive elements
connected in parallel follow a circular trajectory in the admittance plane, clockwise for capacitances,

42
 0.12 0.13 0.14 0.12 0.13 0.14
0.11 0.15 0.11 0.15
0.10 0.38 0.37 0.36 0.10 0.38 0.37 0.36
9 0.39 0.35 0.1 9 0.39 0.35 0.1
0.0 90 6 0.0 90 6
0.40 100 80 0.40 100 80
1 0.3 0.1 1 0.3 0.1
8 70 8 70
0.0 0.4 110 0.9 1 1.2 4 7 0.0 0.4 110 0.9 1 1.2 4 7
0.8 1.4 0 0.8 1.4 0
2
0.4 120 0.7 60 .33 0. 2
0.4 120 0.7 60 .33 0.
07 1.6 18 07 1.6 18
0. 0. 0. 0.
43 0.6 1.8 3 43 0.6 1.8 3
.
0 0 50 2 .
0 0 50 2

0.

0.
06

06
0.2 0.2
13 0.
5 2 13 0.
5 2

19

19
0.

0.
0.

0.
0 44

0 44
31 40

31 40
14 0.

14 0.
0.

0.
05

05
4

4
20

20
0.4 0.4

0.

0.
0.

0.
0.

0.
45

45
30

30
0.

0.
3

3
0.2

0.2
4

4
0.6 0.6

0.0

0.0
0.3

0.3
0.2

0.2
6

6
0

0
1

1
0.4

.4
30

30
15

15
9

9
160 7 0
0.8 0.8

4
0.2

0.2
3

3
0.0

0.0
0.2

0.2
160 7
0.4

0.4
2

2
1 1
0.2

0.2
5

5
8

8
20

20
0.23

0.23
0.02

0.02
0.15

0.15
0.27

0.27
0.48

0.48
0.05 0.1

-0.05 0 0.05 0.1

10

10
170

170
10

10
0.24

0.24
0.01

0.01
0.26

0.26
0.49

0.49
20 50 -50-20 -10

20 50 -50-20 -10
0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9
1.0

1.2

1.4

1.6

1.8
2.0

3.0

4.0

5.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9
1.0

1.2

1.4

1.6

1.8
2.0

3.0

4.0

5.0
0.25
0.25

0.25
0.25
0.00
0.00

0.00
0.00
10

20

50

10

20

50
180

180
0

0
-0.05

0.24

0.24
0.01

0.01
0.26

0.26
0.49

0.49
-170

-170
-10

-10
5 -0.1

-0.1
0.

0.23
0.02

0.02

.15
0.27

0.27
0.48

0.48
-0.1

23

-0
-160

-160
-20

-20
-0.2

-0.2
-5

-5
0.2

0
3

.2
0.0

0.0
1 1

0.2

0.2
7

7
0.4

0.4
2

2
-4

-4
8

8
0.8 0.8
50

50
0.2

0.2
4

4
-3

-3
.3

.3
0.0

0.0
0.2

0.2
6

6
-1

-1
0

0
-0

-0
1

1
0.4

0.4
0.6 0.6

-3

-3
9

9
0.

0.
05

05
20

20
0

0
0.

0.
.4

.4
-4 0.

-4 0.
0.

0.
45

45
06 4

06 4
0.4 0.4
0. -1

-0

0. -1

-0
0 19

0 19
30

30
0.

0.
0.

0.
44

44
.5 -5 -2 .5 -5 -2
30 30

31

31
-0 -0
0.

0.2

0.
-1 7 0. 0 -1
-1 7 0.2 0. 0 -1
0 18 .8 0 18 .8
-0
.6 0. 0. -1
.6 -0
.6 0. 0. -1
.6
20 43 32 0.1 -60 20 43 32 0.1 -60
-1 .08 0. -0.7 7
-1.4 -1 .08 0. -0.7 7
-1.4
0 0
2 -0.8 -0.9 -1
-1.2 0.3 0.1 -70 2 -0.8 -0.9 -1
-1.2 0.3 0.1 -70
9 .4 -110 3 6 9 .4 -110 3 6
0.0 0
-80 0.0 0
-80
1 0.10 -90 -100 0.3 0.15 1 0.10 -90 -100 0.3 0.15
0.4 0.14 4 0.4 0.14 4
0.13 0.12 0.11 0.35 0.13 0.12 0.11 0.35
0.40 0.36 0.40 0.36
0.38 0.39 0.37 0.38 0.39 0.37

Series L Series C
R R

Fig. 50: Circular traces of reactances with varying value connected in series to a fixed impedance

0.12 0.13 0.14 0.12 0.13 0.14

0.11 0.15 0.11 0.15
0.10 0.38 0.37 0.36 0.10 0.38 0.37 0.36
9 0.39 0.35 0.1 9 0.39 0.35 0.1
0.0 90 6 0.0 90 6
0.40 100 80 0.40 100 80
1 0.3 0.1 1 0.3 0.1
8 70 8 70
0.0 0.4 110 0.9 1 1.2 4 7 0.0 0.4 110 0.9 1 1.2 4 7
0.8 1.4 0 0.8 1.4 0
2
0.4 120 0.7 60 .33 0. 2
0.4 120 0.7 60 .33 0.
07 1.6 18 07 1.6 18
0. 0. 0. 0.
43 0.6 1.8 3 43 0.6 1.8 3
0. 0 50 2 0. 0 50 2
0.

0.
06

06

0.2 0.2
13 0.
5 2 13 0.
5 2
19

19
0.

0.
0.

0.
0 44

0 44
31 40

31 40
14 0.

14 0.
0.

0.
05

05
4

4
20

20
0.4 0.4
0.

0.
0.

0.
0.

0.
45

45
30

30
0.

0.
3

3
0.2

0.2
4

0.6 0.6
0.0

0.0
0.3

0.3
0.2

0.2
6

6
0

0
1

1
0.4

.4
30

30
15

15
9

9
160 7 0

0.8 0.8
4

4
0.2

0.2
3

3
0.0

0.0
0.2

0.2
160 7
0.4

0.4
2

2
1 1
0.2

0.2
5

5
8

8
20

20
0.23

0.23
0.02

0.02
0.15

0.15
0.27

0.27
0.48

0.48
0.05 0.1

-0.05 0 0.05 0.1

10

10
170

170
10

10
0.24

0.24
0.01

0.01
0.26

0.26
0.49

0.49
20 50 -50-20 -10

20 50 -50-20 -10
0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9
1.0

1.2

1.4

1.6

1.8
2.0

3.0

4.0

5.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9
1.0

1.2

1.4

1.6

1.8
2.0

3.0

4.0

5.0
0.25
0.25

0.25
0.25
0.00
0.00

0.00
0.00
10

20

50

10

20

50
180

180
0

0
-0.05

0.24

0.24
0.01

0.01
0.26

0.26
0.49

0.49
-170

-170
-10

-10
5 -0.1

-0.1
0.

0.23
0.02

0.02

.15
0.27

0.27
0.48

0.48
-0.1

23

-0
-160

-160
-20

-20
-0.2

-0.2
-5

-5
0.2

0
3

.2
0.0

0.0

1 1
0.2

0.2
7

7
0.4

0.4
2

2
-4

-4
8

0.8 0.8
50

50
0.2

0.2
4

4
-3

-3
.3

.3
0.0

0.0
0.2

0.2
6

6
-1

-1
0

0
-0

-0
1

1
0.4

0.4

0.6 0.6
-3

-3
9

9
0.

0.
05

05
20

20
0

0
0.

0.
.4

.4
-4 0.

-4 0.
0.

0.
45

45
06 4

06 4

0.4 0.4
0. -1

-0

0. -1

-0
0 19

0 19
30

30
0.

0.
0.

0.
44

44

.5 -5 -2 .5 -5 -2
30 30
31

31

-0 -0
0.

0.2
0.

-1 7 0. 0 -1
-1 7 0.2 0. 0 -1
0 18 .8 0 18 .8
-0
.6 0. 0. -1
.6 -0
.6 0. 0. -1
.6
20 43 32 0.1 -60 20 43 32 0.1 -60
-1 .08 0. -0.7 7
-1.4 -1 .08 0. -0.7 7
-1.4
0 0
2 -0.8 -0.9 -1
-1.2 0.3 0.1 -70 2 -0.8 -0.9 -1
-1.2 0.3 0.1 -70
9 -110 3 9 -110 3
0.0 0.4 6 0.0 0.4 6
-100 -80 0.3 0.15 -100 -80 0.3 0.15
0.4
1 0.10 -90
0.14 4 0.4
1 0.10 -90
0.14 4
0.13 0.12 0.11 0.35 0.13 0.12 0.11 0.35
0.40 0.36 0.40 0.36
0.38 0.39 0.37 0.38 0.39 0.37

Shunt L R Shunt C R

Fig. 51: Circular traces of reactances with varying value connected in parallel to a fixed impedance

anticlockwise for inductances.

This rule is illustrated in Fig. 52.

8.3.3 Impedance transformation using a transmission-line

The S-matrix of an ideal, lossless transmission-line of physical length l is given by
[ ]
0 e−jβl
S= , (53)
e−jβl 0

43
 0.12 0.13 0.14
0.11 0.1
0 5
0.1 0.38 0.37 0.36
9 0.39 0.3 0.1
0.0 0.4
0
100
90 80 5 6
1 0.3 0.
08 70
0. 0.4 110 0.9 1 1.2 4 17
0.8 1.4 0
42
0. 120 0.7 60 .33 0.
07 1.6 18
0. 6 0.
43 0. 1. 3
0. 0 8 50 2

0.
06
0.2
13 0.
5 2

19
0.

0.
0 44
Shunt L

31 40
14 0.

0.
05

20
0.4

Series L

0.
0.

0.
45

30
0.

0.
04
0.6

21
3

0.
6
0.

0
0.
.4

30
15

29
160 7 0
0.8

0.2
3
0.0

0.2
0.4

2
1

0.2

8
20

0.23
0.02

0.27
0.48

0.1
Series C
Shunt C

0.1

10
170

10

0.24
0.01

0.26
0.49

20 50 -50-20 -10
0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9
1.0

1.2

1.4

1.6

1.8
2.0

3.0

4.0

5.0

0.25
0.25
0.00
0.00

10

20

50
180
0

0
-0. 05

0.24
0.01

0.2
0.49

70

-10
5 -0.1

6
-1

0.23
0.02

0.27
0.48

.1
2 -0
0.0 0

-
6

2
-5
1

-0.

0.2
3

0
-

0.2
7
0.4

2
-4

8
0.8
50

0
04

-3
.3

.2

0.
46

-1

0
0.

1
0.6

29
-
0.

-3

0.
05

20
0
0.

.4

-4 0.

0.
45

0.4
0. -1

-0

0 19

30
0.

06

0.
44

.5 -5 -2
30

31
-0
0.

-1 7 0.2 0. 0 -1
0 18 .8
-0
.6 0. 0. -1
.6
20 43 32 0. -60
-1 .08 0. -0.7 17 -1.4
0
-0.8 -1.2 0.
-11
0 9 42 -0.9 -1 33 0.1 -70
0.0 0. 6
0 -100 -80 0.3 0.1
0.1 1 -90 5
0.4 0.11 0.14 4
0 0.13 0.12 0.3
0.4 5
0.39 0.36
0.37 0.38

Fig. 52: Navigation in the Smith chart when connecting reactive elements.

where β = 2π/λ is the propagation coefficient at the wavelength λ (λ = λ0 for ϵr = 1).

The lossless transmission-line changes only the phase between its ports. Adding a short piece of,
e.g. of coaxial cable in front of a load impedance, will turn the corresponding circle of Zload clockwise,
which is effectively a transformation of the reflection factor Γload (without line) to the new reflection
factor Γin = Γload e−j2βl . Graphically speaking, the vector corresponding to Γin is rotated clockwise by
an angle of 2βl (Fig. 53).
0.12 0.13 0.14
0.11 0.1
0 5
0.1 0.38 0.37 0.36
9 0.39 0.3 0.1
0.0 0.4
0
100
90 80 5 6
1 0.3 0.
08 70
0. 0.4 110 0.9 1 1.2 4 17
0.8 0

0.
07
43
42
0. 120
0.
6
0.7
Γload 1.4
1.6
60 .33

1.
0.
3
0.
18

0. 0 8 50 2
0.
06

0.2
13 0.
5 2
19
0.

0.
0 44

31 40
14 0.

0.
05

20

0.4
0.
0.

0.
45

30
0.

0.
04

0.6
21
3

0.
6
0.

0
0.
.4

30
15

29
160 7 0

0.8
4

0.2
3

2βl
0.0

0.2
0.4

1
.2

8
20
5 0

0.23
0.02

0.27
0.48

1 0.1

10
170

10
0 0.05 0.

0.24
0.01

0.26
0.49

20 50 -50-20 -10
0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9
1.0

1.2

1.4

1.6

1.8
2.0

3.0

4.0

5.0

0.25
0.25
0.00
0.00

10

20

50
180

0
-0.05

Γin
0.24
0.01

0.26
0.49

-170

-10
5 -0.1

0.23
0.02

0.27
0.48

-0.1
0.0 0

-2
-16

-0.2

-5

0.2
3

1
0.2
7
0.4

2
-4

0.8
50

0
04

-3
.3

.2

0.
46

-1

0
0.

-0

0.6
29
0.

-3

0.
05

20
0
0.

.4

-4 0.

0.
45

06 4

0.4
0. -1

-0

0 19

30
0.

0.
44

.5 -5 -2
30
31

-0
0.

-1 7 0.2 0. 0 -1
0 18 .8
-0
.6 0. 0. -1
.6
20 43 32 0. -60
-1 .08 0. -0.7 17 -1.4
0
-0.8 -1.2 0.
9 42 -11
0 -0.9 -1 33 0.1 -70
0.0 0. 6
0 -100 -80 0.3 0.1
1 0.1 -90 5
0.4 0.11 0.14 4
0 0.13 0.12 0.3
0.4 5
0.39 0.36
0.37 0.38

Fig. 53: Adding a lossless transmission-line of physical length l to an impedance Zload

The input impedance of a lossless transmission-line of characteristic impedance Z0 , terminated

with Zload is given by:
Zload + jZ0 tan(βl)
Zin = Z0 (54)
Z0 + jZload tan(βl)

44
 Im (Z)

inductive
λ λ
4 2
Re (Z)

capacitive

Fig. 54: Impedance of a transmission line as a function of its length l

and the corresponding reflection coefficient follows as mentioned:

Γin = Γload e−j2βl (55)
Depending on the values of β, Z0 , Zload , and l, the input impedance will be quite different from the load
impedance Zload . Special cases are:
– l = λ/2: Zin = Zload
– l = λ/4: Zin = Z02 /Zload (impedance transformer)
– Zload = Z0 : Zin = Z0 (matched termination)
– Zload = jXload : Zin = jXin (reactive load ⇒ reactive input impedance)
– l ≪ λ: Zin = Zload (basically no line present)
Terminating a transmission-line with a short circuit, Zload = 0, simplifies Eq. 54 to
Zin = jZ0 tan(βl) (56)
which results in an “inductive” or “capacitive” impedance behavior at the input, depending on the length
of the line (see Fig. 54).
Adding a transmission-line of length λ/4 interestingly results in a change of Γ by a factor −1:
l= λ
Γin = Γload e−j2βl = Γload e−j2( λ )l =4 Γload e−jπ = −Γload .
2π
(57)
Again, this is equivalent to inverting an impedance z to its admittance 1/z, or the clockwise rotation of
the impedance vector by 180◦ . Especially when starting with a short circuit (Zload = 0 ⇒ −1 in the
Smith chart), adding a transmission line of length λ/4 transforms it into an open circuit (+1 in the Smith
chart), and vice versa.

8.3.4 Two-port examples

The general form of Eq. 55 returns the input reflection coefficient Γin for a 2-port network terminated
with Zload , i.e. a reflection coefficient Γout at the output port:
S12 S21 Γload
Γin = S11 + . (58)
1 − S22 Γload
Lets evaluate some examples, defined by their S-matrix, which map their impedance to particular charac-
teristic lines and circles on the Smith chart. For this illustration, a very simplified Smith chart, consisting
just of the outermost circle (imaginary axis) and the real axis is used.

45
 z=0
Z
lo
ad
in
cr z=1
e as
in
g

z=∞

Fig. 55: Rotation of the real axis, therefore the reference plane of the Smith chart when adding a
transmission-line

8.3.4.1 Transmission-line of length λ/16

The S-matrix of a λ/16 transmission-line is
[ ]
e−j 8
π
0
S= −j π8 (59)
e 0
has a input reflection coefficient of
Γin = Γload e−j 4
π
(60)
This corresponds to a rotation of the real axis of the Smith chart by an angle of 45◦ (Fig. 55) and
hence a change of the reference plane of the chart (Fig. 55). Consider, for example, a transmission-
line terminated by a short and hence Γload = −1. The resulting reflection coefficient is then equal to
Γin = e−j 4 .
π

8.3.4.2 3 dB attenuator
The S-matrix of a 3 dB attenuator is given by
[ √ ]
2
0
√ 2
S= 2
. (61)
2 0

The resulting reflection coefficient is

Γload
Γin = (62)
2

z=0 z=1 z=∞

Fig. 56: Effect of an attenuator in the Smith chart

46
 z=0 z=∞
z=1

Fig. 57: A load resistor of variable value in the simplified Smith chart. Since the impedance has a real
part only, the trace remains on the real axis of the Γ plane.

In the Smith chart, the connection of such an attenuator causes the outermost circle to shrink to a radius
of 0.5, see Fig. 5610 .

8.3.5 Resistive load

Fig. 57 illustrates how the real axis is passed, if a resistive load changes its value 0 < z < ∞.

8.4 Examples for applications of the Smith chart

In this section two examples of typical RF problems demonstrate how the Smith chart greatly facilitates
their solutions.

8.4.1 A step in the characteristic impedance

Consider a junction between two infinitely short cables, an incoming with a characteristic impedance
of Z1 = 50Ω, the outgoing with Z2 = 75Ω (Fig. 58). Both ports are matched in their characteristic
impedance.
The incident waves are denoted with ai (i = 1, 2), the reflecting waves with bi . The reflection
coefficient at port 1 follows as
Z2 − Z1 75 − 50
Γ1 = = = +0.2. (63)
Z2 + Z1 75 + 50
10
An attenuation of 3 dB corresponds to a reduction by a factor 2 in power.

a1 a2
Junction between a
50 Ω and a 75 Ω cable
(infinitely short cables)
b1 b2

Fig. 58: Junction between two coaxial cables, one with with Z1 = 50Ω, the other with Z2 = 75Ω
characteristic impedance. Infinitely short cables are assumed – only the junction is considered.

47
 0.12 0.13 0.14
0.11 0.1
0 5
0.1 0.38 0.37 0.36
9 0.39 0.3 0.1
0.0 0.4
0
100
90 80 5 6
1 0.3 0.
08 70
0. 0.4 110 0.9 1 1.2 4 17
0.8 1.4 0
42
0. 120 0.7 60 .33 0.
07 1.6 18
0. 6 0.
43 0. 1. 3
0. 0 8 50 2

0.
06
0.2
13 0.
5 2

19
0.

0.
0 44

31 40
14 0.

0.
05

20
0.4

0.
0.

0.
45

30
0.

0.
04
0.6

21
3

0.
6
0.

0
0.
.4

30
15

29
160 7 0
0.8

0.2
3
0.0

0.28
0.4

2
1

.2

20
5 0
V1 ∝ a1 + b1 = 1.2

0.23
0.02

0.27
0.48

1 0.1

10
170

10
0 0.05 0.

0.24
0.01

0.26
0.49

20 50 -50-20 -10
0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9
1.0

1.2

1.4

1.6

1.8
2.0

3.0

4.0

5.0

0.25
0.25
0.00
0.00

10

20

50
180

0
-0.05

0.24
0.01

I1 Z ∝ a1 − b1 −b1 b1 = +0.2

0.26
0.49

-170

-10
5 -0.1

0.23
0.02

0.27
0.48

-0.1
0.0 0

-2
-16

-0.2

-5

0.2
3

0
1

0.2
7

Incident wave a1 = 1
0.4

2
-4

8
0.8
50

0.
04

-3
3

0.
46

05 -1
.

0
0.

-0

1
0.6

29
0.

-3

0.
20
0
0.

.4

-4 0.

0
45

06 4

0.4

.
0. -1

-0

0 19

30
0.

0.
4

.5 -5 -2
4

30

31
-0
0.

-1 7 0.2 0. 0 -1
0 18 .8
-0
.6 0. 0. -1
.6
20 43 32 0. -60
-0.7 -1 .08 0. 17 -1.4
0
0 2 -0.8 -0.9 -1
-1.2 0.
33 0.1 -70
1 9 4
-1 0.0 0. 6
0 -100 -80 0.3 0.1
0.1 1 -90 5
0.4 0.11 0.14 4
0 0.13 0.12 0.3
0.4 5
0.39 0.36
0.37 0.38

Fig. 59: Visualization of the two-port formed by the two cables of different characteristic impedances

Thus, the voltage of the reflected wave at port 1 is 20% of the incident wave (b1 = a1 · 0.2), and the
reflected power at port 1 is Γ21 = 0.04 ≡ 4%. From conservation of energy, the transmitted power has to
be 96%, i.e. b22 = 1 − Γ21 = 0.96.
The voltage transmission coefficient in this particular case computes t = 1 + Γ, and the output
voltage of the transmitted wave at port 2 is higher than the voltage of the incident wave at port 1:
Vtransmitted = Vincident + Vreflected = 1 + 0.2 = 1.2. Also,√note that this structure is not symmetric
(S11 = +0.2 ̸= S22 = −0.2), but reciprocal (S21 = S12 = 1 − Γ21 ). As all impedances are real, the
corresponding vectors show up in the Smith chart on the real axis (Fig. 59).

8.4.2 Quality (Q) factor of a cavity

The second example shows the calculation of the quality factor of a cavity resonator with help of the
Smith chart.
A cavity at or near to one of its eigenmode resonances can be approximated by a parallel RLC
equivalent circuit (Fig. 60). The resonance condition is given as

Zinput Zshunt

ZG
R L C Vbeam
V0

Fig. 60: Equivalent circuit of a cavity near resonance. The transformer describes the coupling of the
cavity (typically Zshunt ≈ 1 MΩ, as seen by the beam) to the generator (often ZG = 50 Ω).

48
 −
Im (Z) f = f(−3dB)

f =0
45◦ f = f(res)

Re (Z)
f →∞

+
f = f(−3dB)

Fig. 61: Schematic drawing of the 3 dB bandwidth in the impedance plane

1
ωL = (64)
ωC
from which the resonance frequency follows
1 1 1
ωres = √ or fres = √ . (65)
LC 2π LC

The impedance Z of the cavity equivalent circuit is simply

1
Z(ω) = 1 1 . (66)
R + jωC + jωL

The 3 dB bandwidth ∆f refers to the points where Re(Z) = Im(Z), which correspond√ to two
◦
vectors with an argument of 45 (Fig. 61) and an impedance of |Z(−3 dB) | = 0.707R = R/ 2.
In general, the quality factor Q of a resonant circuit is defined as the ratio of the stored energy W
over the energy dissipated P in one oscillation cycle:
ωW
Q= . (67)
P
However, the Q factor for a resonance can also be calculated using the 3 dB bandwidth and the resonance
frequency:
fres
Q= . (68)
∆f
For a cavity, three different quality factors are defined:

– Q0 (unloaded Q): Q factor of the unperturbed system, i.e. the stand-alone cavity;
– QL (loaded Q): Q factor of the cavity when connected to a generator and/or measurement circuits;
– Qext (external Q): Q factor that describes the degeneration of Q0 due to the generator and/or
diagnostic impedances.

All these Q factors are linked via a simple relation:

1 1 1
= + . (69)
QL Q0 Qext

49
 0.13

9
0.1
0
0.11
0.39
0.12
0.38 Locus of Im (Z) = Re (Z)
0.37
0.14
0.36
0.1
0.3
5
0.1
0.0 0.4
0
100
90 80 5 6
1 0.3 0.
08 70
0. 0.4 110 0.9 1 1.2 4 17
0.8 1.4 0
42
0. 120 0.7 60 .33 0.
07 1.6 18
0. 6 0.
43 0. 1. 3
0. 0 8 50 2

f1

0.
06
0.2
13 0.
5 2

19
0.

0.
0 44

31 40
14 0.

0.
05
f3

20
0.4

0.
0.

0.
45

30
0.

0.
04
0.6

f5

21
3

0.
6
0.

0
0.
.4

30
15

29
160 7 0
0.8

0.2
3
0.0

0.2
0.4

2
1

0.2

8
20

0.23
0.02

0.27
0.48

0.1
5 0.1

10
170

10

0.24
0.01

0.26
0.49

20 50 -50-20 -10
0.0
f0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9
1.0

1.2

1.4

1.6

1.8
2.0

3.0

4.0

5.0

0.25
0.25
0.00
0.00

10

20

50
180
0

0
-0.05

0.24
0.01

0.2
0.49

70

-10
5 -0.1

6
-1

0.23
0.02

0.27
0.48

-0.1
0.0 0

-20
6

-0.2

-5
-1

0.2
3

0.2
7
0.4

2
-4

8
f6
0.8
0

0
04

-
5 -15

30
. 21

0.
46

0.
0.

0.6

29
-
0.

-3
f4

0.
0

20
6 40
0.

.4

-4 0.

0
45

0.4

.3
0. -1

-0

0 19
0.

0
f2
0

0.
44

.5 -5 -2
30

31
-0
0.

-1 7 0.2 0. 0 -1
0 18 .8
-0
.6 0. 0. -1
.6
20 8 43 32 0. -60
-0.7 - 1 0 0. 17 -1.4
0.
-0.8 -1.2 0.
-11
0 9 42 -0.9 -1 33 0.1 -70
0.0 0. 6
0 -100 -80 0.3 0.1
0.1 1 -90 5
0.4 0.11 0.14 4
0 0.13 0.12 0.3
0.4 5
0.39 0.36
0.37 0.38

Fig. 62: Evaluation of the different Q factors of a resonant cavity with help of the Smith chart

The coupling coefficient β is then defined as

Q0
β= . (70)
Qext
This coupling coefficient has not to be confused with the propagation coefficient of transmission lines,
which is also denoted as β.
In the Smith chart, a resonant circuit shows up as a circle (Fig. 62, dashed, red circle shown in the
“detuned short” position). The larger the circle, the stronger is the coupling. Three types of coupling are
distinguished, depending on the range of beta (= size of the circle, assuming the circle is in the “detuned
short” position):

– Undercritical coupling (0 < β < 1): the radius of the resonance circle is smaller than 0.25. Hence,
the center of the chart (Γ = 0) lies outside the circle.
– Critical coupling (β = 1): the radius of the resonance circle is exactly 0.25. Hence, the circle
crosses Γ = 0 at the resonance frequency fres .
– Overcritical coupling (1 < β < ∞): the radius of the resonance circle is larger than 0.25. Hence,
the center of the chart lies inside the circle.

In practice, the circle may be rotated around the origin due to the transmission lines between the resonant
circuit and the measurement device.
From the different marked frequency points in Fig. 62 the 3 dB bandwidth, and thus the quality
factors Q0 , QL and Qext are determined as follows:

– The unloaded Q is determined from f5 and f6 . The condition for these points is Re(Z) = Im(Z),
with the resonance circle in the “detuned short” position.
– The loaded Q is determined from f1 and f2 . The condition to find these points is |Im(S11 )| →
max. in “detuned short” position.
– The external Q is calculated from f3 and f4 . The condition to determine these points is Z = ±j in
“open short” position, which is equivalent to Y = ±j in “detuned short” position

To determine the points f1 to f6 with a network analyzer, the following steps are applicable:

50
 – f1 and f2 : set the marker format to Re(S11 ) + j Im(S11 ) and determine the two points where
Im(S11 ) = max.
– f3 and f4 : set the marker format to Z and find the two points where Z = ± j.
– f5 and f6 : set the marker format to Z and locate the two points where Re(Z) = Im(Z).

9 Summary
Some fundamental concepts on RF devices, instruments, and singal processing techniques have been
presented in this introduction to RF measurement concepts. Advantages of various measurement methods
using spectrum and network analyzers were presented. In the last section the definition of the Smith chart,
and its usage were illustrated with several examples. This article supports the practical part of the CAS
intermediate-level RF course, and serves as background information.

10 Hand-on experiments
The following hands-on experiments are foreseen:

10.1 Spectrum analyzer test stand 1:

– Measurements of several types of modulation (AM, FM and PM) in time and frequency domain.
– Superposition of AM and FM spectra (unequal carrier sidebands).
– Concept of a spectrum analyzer: the superheterodyne method. Practice different settings (video
bandwidth, resolution bandwidth etc). Advantage of FFT spectrum analyzers.

10.2 Spectrum analyzer test stand 2:

– Measurement of the TOI point of some amplifiers (intermodulation tests).

– Concept of noise-figure and noise-temperature measurements, testing a noise diode, the basics of
thermal noise.
– Electromagnetic compatibility (EMC) measurements (e.g. analyze your cell-phone spectrum).
– General concepts of non-linear distortions and application of vector spectrum analyzers, spectro-
gram mode.
– Measurement of the RF characteristic of a microwave detector diode (output voltage versus input
power ... transition between regimes output voltage proportional to input power and output voltage
proportional to input voltage).

10.3 Spectrum analyser test stand 3:

– Concept of noise-figure and noise-temperature measurements, testing a noise diode, the basics of
thermal noise.
– Noise-figure measurements on amplifiers and also attenuators.
– The concept and meaning of excess noise ratio (ENR) numbers.
– Noise temperature of the fluorescent tubes in the room using a satellite receiver.

51
10.4 Network analyser test stand 1:

– Calibration of the vector network analyzer.

– Navigation in the Smith chart.
– Application of the triple-stub tuner for matching.
– Measurements of the light velocity using a trombone (constant-impedance adjustable coaxial line)
in the frequency domain.
– N -port (N = 1–4) S-parameter measurements for different reciprocal and non-reciprocal RF
components.
– Self-made RF components: calculate, build and test your own attenuator (and then take it home).

10.5 Network analyzer test stand 2:

– Measurements of the light velocity using a trombone (constant-impedance adjustable coaxial line)
in the time domain.
– Two-port measurements for active RF components (amplifiers).
– A 1 dB compression point (power sweep).
– Beam transfer impedance measurements with the wire (button pick-up, stripline pick-up).

10.6 Network analyzer test stand 3:

– Measurements of the characteristic cavity features (Smith-chart analysis).

– Cavity perturbation measurements (bead pull).
– Perturbation measurements using rectangular waveguides.
– Standing wave ratio (SWR) measurements using a waveguide measurement line and movable
probe.

References
[1] G.D. Vendelin, A.M. Pavio and U.L. Rohde, Microwave Circuit Design Using Linear and Nonlin-
ear Techniques, second ed. (Wiley-Interscience, New Jersey, 2005), ISBN-10 0-471-41479-4.
[2] F. Caspers, Proc. CERN Accelerator School, RF Engineering for Particle Accelerators, Oxford,
UK, 1991, p.181.
[3] M. Thumm, W. Wiesbeck and S. Kern, Hochfrequenzmesstechnik (Teubner, Stuttgart/Leipzig,
1998), ISBN 3-519-16360-8.
[4] R.A. Witte, Spectrum and Network Measurements (Prentice-Hall, New Jersey , 1991), ISBN 0-13-
826959-9.
[5] W.O. Schleifer, Hochfrequenz und Mikrowellenmesstechnik in der Praxis (Hüthig, Heidelberg,
1981), ISBN 3-7785-0675-7.
[6] B. Schiek and H.J. Sieveris, Rauschen im Hochfrequenzschaltungen (Hüthig, Heidelberg, 1984),
ISBN 3-7785-2007-5.
[7] P.C.L. Yip, High Frequency Circuit Design and Measurement (Chapman and Hall, London, 1990),
ISBN 0-412-34160-3.
[8] G. Evans and C.W. McLeisch, RF-Radiometer Handbook (Artech, Dedham, 1977), ISBN 0-89006-
055-X.
[9] F.R. Connor, Noise (Edward Arnold, London, 1973), ISBN 0-7131-3306-6.

52
[10] F. Landstorfer and H. Graf, Rauschprobleme der Nachrichtentechnik (Oldenbourg, München,
1981), ISBN 3-486-24681-X.
[11] O. Zinke and H. Brunswig, Lehrbuch der Hochfrequenztechnik, Zweiter Band (Springer, Berlin,
1974), ISBN 3-540-06245-9.
[12] Agilent Technologies, Inc., Fundamentals of RF and microwave noise figure measurements, Agilent
Application Note 57-1, 2010.
[13] B. Schiek, Messysteme der Hochfrequenztechnik (Hüthig, Heidelberg, 1984), ISBN 3-7785-1045-2.
[14] F. Caspers, RF engineering basic concepts: S-parameters, CAS Proc., 2010, CERN Yellow Report
CERN-2011-007, pp. 67-93.
[15] J. Verspecht and D. Root, Polyharmonic Distortion Modeling, IEEE Microwave Magazine, Vol. 7,
Issue 3, June 2006, pp. 44-57.
[16] M. Thumm, W. Wiesbeck and S. Kern, Hochfrequenzmesstechnik, Verfahren und Messsysteme
(Teubner, Stuttgart, 1998), ISBN 978-3519163602.
[17] M. Hiebel, Fundamentals of Vector Network Analysis (Rohde & Schwarz, München, 2007), ISBN
3939837067.
[18] Agilent Technologies, Inc., Understanding the fundamental principles of vector network analysis,
Agilent Application Note AN 1287-1, 2000.
[19] Anritsu Company, Time domain measurements using vector network analyzers, Anritsu Application
Note No. 11410-00206, R , 2009.
[20] Agilent Technologies, Inc., Time domain analysis using a network analyzer, Agilent Application
Note 1287-12, 2012.
[21] H. Meinke and F.-W. Gundlach, Taschenbuch der Hochfrequenztechnik (Springer, Berlin, 1992).
[22] P. Smith, Electronic Applications of the Smith Chart (Noble Publishing, Atlanta, 2000), ISBN 1-
884932-39-8.

11 Appendix: Meaning of the rulers below the Smith chart

How to use the rulers that are often plotted below the Smith chart?
A commonly used set of rulers is usually found below the Smith chart, see Fig. 63. There are four
rulers, some with an upper and lower part, to quickly estimate and compare some important properties in
terms of modulus values. For the following discussion lets split the upper three rulers at the line marked
CENTER to a left and right part, each to be discussed separately. These rulers start at the CENTER,
referring to the center of the Smith chart, and end at the left or right boundary, referring to the circular
boundary of the Smith chart. The 4th ruler at the bottom is different, it starts at the left boundary ORIGIN
and ends at the right boundary.

Fig. 63: Example for a set of rulers that can be found underneath the Smith chart (please note corrections
in respect to the RF-course printouts)

First ruler, left/upper part in Fig. 64 is marked as SWR which mean actually VSWR, i.e. voltage
standing wave ratio. It ranges between one – for the matched case (center of the Smith chart) and infinity
– for total reflection (boundary of the Smith chart), respectively. The upper part is in linear scale, the

53
 Fig. 64: Left part of the rulers usually plotted underneath the Smith Chart

lower part of this ruler is in dB, noted as dBS (dB referred to Standing Wave Ratio). Example: SWR =
10 corresponds to 20 dBS, SWR = 100 corresponds to 40 dBS [voltage ratios, not power ratios].
Second ruler, left/upper part, marked as RTN.LOSS i.e. return loss in dB. This indicates the amount
of reflected wave expressed in dB. Thus, in the center of SC nothing is reflected and the return loss is
infinite. At the boundary we have full reflection, thus return loss is 0 dB. The lower part of the scale
denoted as RFL.COEFF.P is a reflection coefficient in terms of POWER (proportional |Γ|2 ). If there is
no reflected power for the matched case locus is in the center of the Smith chart (SC). On the contrary, if
normalized reflected power is equal to 1 locus is at the boundary.
Third ruler, left, marked as RFL.COEFF,E or I gives us the absolute value of the reflection coeffi-
cient in linear scale. Note that since we have the modulus we can refer it both to voltage or current as we
have omitted the sign. Obviously in the center the reflection coefficient is zero, at the boundary it is one.
The fourth is a Voltage transmission coefficient. Note that the modulus of the voltage (and current)
transmission coefficient has a range from zero, i.e. short circuit, to +2 (open = 1+Γ with Γ = 1). This
ruler is only valid for Zload = real, i.e. the case of a step in characteristic impedance of the coaxial line.
Third ruler, right (see Fig. 65) marked as TRANSM.COEFF.P refers to the transmitted power as a
function of mismatch and displays essentially the relation Pt = 1 − |Γ|2 . Thus, in the center of the SC
full match, all the power is transmitted. At the boundary we have total reflection and e.g. for a Γ value
of 0.5 we see that 75 % of the incident power is transmitted.

Fig. 65: Right part of the rulers usually plotted underneath the Smith Chart

Second ruler, right/upper part, denoted as RFL.LOSS in dB denotes reflection loss. This ruler
refers to the loss in the transmitted wave, and should not be confounded with the return loss referring to
the reflected wave. It displays the relation Pt = 1 − |Γ|2 in dB. This ruler is nowadays rather not more

54
in use.
√
Let us analyse an example from Fig. 66: |Γ| = 1/ 2 = 0.707 , transmitted power = 50 % thus
loss = 50 % = 3 dB. Note√that in the lowest ruler the voltage of the transmitted wave (Zload = real) would
be Vt = 1.707 = 1 + 1/ 2 if referring to the voltage.

√
Fig. 66: Example for |Γ| = 1/ 2 = 0.707 and 50 % of transmitted power (i.e. 3 dB loss), see description
in text

Finally, the First ruler, right/upper part, denoted as ATTEN. in dB assumes that one is measuring
an attenuator or a lossy line which itself is terminated by an open or short circuit (full reflection). Thus
the wave is traveling twice through the attenuator (forward and backward). The value of this attenuator
can be between zero and some very high number corresponding to the matched case. The lower scale of
first ruler displays the same situation just in terms of VSWR.
For the next example see Fig. 67: an 10 dB attenuator attenuates the reflected wave by 20 dB going
forth and back and we get a reflection coefficient of Γ = 0.1. This correspond to the reflection of 10 %
in voltage. Another example is 3 dB attenuator: for the forth and back transmission it gives 6 dB which
correspond to half of the voltage. Table 5 is reprinted from an original paper of Phillip H. Smith [22] and
summarizes reflection formulas discussed above.

Fig. 67: Example for 10 dB and 3 dB attenuator, see description in text

55
 Table 5: Reflection formulas
function traveling waves reflection coefficient standing waves

r S−1
VOLTAGE REFL. COEFF. i Γ S+1

POWER REFL. COEF. ( ri )2 Γ2 ( S−1

S+1 )
2

RETURN LOSS [dB] 10 · log( ri )2 −10 · log(Γ2 ) −10 · log( S−1

S+1 )
2

2
REFLECTION LOSS [dB] 10 · log( i2 −r
i
2) −10 · log(1 − Γ2 ) −10 · log[1 − ( S−1 2
S+1 ) ]

[(i+r)/(i−r)]2 1−Γ+Γ2 −Γ3 1+S 2

STDG. WAVE LOSS COEF. 1− 2[(i+r)/(i−r)] 1−Γ−Γ2 +Γ3 2S

STDG. WAVE RATIO [dB] 20 · log( i+r

i−r ) 20 · log( 1+Γ
1−Γ ) 20 · log(S)

√
MAX. OF STDG. WAVE ( i+r
i−r )
1/2 ( 1+Γ
1−Γ )
1/2 S

MIN. OF STDG. WAVE ( i−r

i+r )
1/2 ( 1−Γ
1+Γ )
1/2 √1
S

i+r 1+Γ
STANDING WAVE RATIO i−r 1−Γ S

ATTENUATION [dB] −10 · log( ri ) −10 · log(Γ) −10 · log( S+1

S−1
)

whereas: i = incident wave amplitude, r = reflected wave amplitude, Γ = reflection coefficient,

S ≡ SWR = voltage standing wave ratio.

56