Engineering for Advanced Radar and

Electromagnetic Spectrum Operations

NEW ERA OF SECURITY
Preface
Defense forces in the modern world are progressing toward a new era of electromagnetic spectrum
operations (EMSO). The availability and rapid evolution of technology results in a conflict domain filled with
advanced threats and foes. Military forces strive to excel in this spectrum across all domains of land, sea,
air and space. The electromagnetic (EM) spectrum is used to detect, deceive, and disrupt the efforts of
any unwanted threat activity.

Electronic warfare (EW) generally includes anything operating in the radio frequency (RF). EM
operations target many aspects of the EW environment, ranging from radars and jammers to military
communications. EW systems use the EM spectrum to support communications, sensing, and defense.
Modern EW systems continuously evolve as new and emerging technologies transform these systems.
Digital and programmable RF equipment, such as software-defined radios, adds complexity. In addition,
radars can quickly change waveforms, making it challenging to locate, identify, and confuse hostile
emitters. These trends impact every aspect of EW. You can address these evolving needs with flexible,
scalable threat simulation and analysis solutions.

This book outlines the core aspects of radar and EMSO engineering. Starting with broad fundamentals,
it covers the technology shift from analog systems to faster digital systems. From the operational
perspective, these systems deploy wideband frequencies. Signal formats and modulation schemes —
pulsed and otherwise — continue to grow more complicated, and this demands wider bandwidth.
Advanced digital signal processing (DSP) techniques are used to disguise system operation and
avoid jamming.

This volume explores all possible challenges and scenarios at the design and simulation, component,
subsystem, and system levels. The design section highlights the importance of model-based
engineering and the significance of simulation tools, which can help increase accuracy at much lower
cost. The subsequent sections take a bottoms-up approach, covering critical amplifier measurements,
material measurement, filters, mixers, and up/down converters. At the subsystem level, the transmitter
and receiver sections highlight how the latest signal generators and signal analyzers can help
engineers perform validation in their labs using software. The last section provides a complete overview
of the system level, addressing the importance of software intelligence, data analysis, and emerging
cyber needs.

Offering a combination of leading technical expertise and a heritage in measurement science and
innovation, Keysight works closely with research, design, and manufacturing companies to support the
development and evolution of new technologies. We offer solutions to address all phases of development
and support for radar and EW systems, from design to verification through long-term sustainment. At
Keysight, we take pride in our leadership as a commercial collaborator, creating and delivering the rapidly
adaptable solutions you need for success in your current programs and to achieve your objectives for
future capabilities.
Table of Contents

Section 1: Prepare for a New Era of Security Threats

Chapter 1 - Introduction & Evolving Challenges in Radar and EW................................................ 6

Section 2: Virtual Prototyping using Design Tools

Chapter 2 - Radar and EW Development Using Model-Based Engineering..................................16

Chapter 3 - Radar and EW System Design and Interference Analysis.........................................26

Chapter 4 - Simulation and Verification of Pulse Doppler Radar Systems...................................41

Chapter 5 - Applying Ultra-Wideband Technology in Radar and EW Systems..............................57

Chapter 6 - Dealing with the Complexity of Phased-Array Systems............................................63

Chapter 7 - Generating Signals for Radar and EW Design and Verification..................................75

Chapter 8 - Virtual Flight Testing of Radar System Performance.................................................88

Section 3: RF Component Test and Verification

Chapter 9 - Figures of Merit Aid Amplifier Distortion Assessment...............................................96

Chapter 10 - Determine Your Optimal Noise Figure Measurement Method................................101

Chapter 11 - Select the Appropriate Phase Noise Test Solution................................................106

Chapter 12 - Measure S-Parameters with Increased Confidence.............................................119

Chapter 13 - Characterizing Frequency Translating Devices....................................................135

Chapter 14 - Expand Insight into Nonlinear Behavior and Modeling.........................................145

Chapter 15 - Methods for Measuring the Dielectric Properties of Materials..............................148

Section 4: Radar and EW Subsystem Test

Chapter 16 - Signal Source Essentials- Parameters, Technologies, and Methods......................159

Chapter 17 - Measuring Radar and EW Signals.......................................................................182

Chapter 18 - Testing TR Modules Efficiently............................................................................208

Chapter 19 - Understanding the Techniques for Antenna Testing..............................................222

Section 5: System Level Radar and EW Test

Chapter 20 - Avoiding 5G Coexistence and Effectively Simulating Background Signals.............236

Chapter 21 - Validating Jammer Effectiveness........................................................................245

Chapter 22 - Radar and EW Signal Recording and Analysis.....................................................251

Chapter 23 - Electronic Intelligence.......................................................................................275

Chapter 24 - Test System Calibration and Alignment...............................................................279

Chapter 25 - Field Testing, Sustainment, Uptime-Service, and Support....................................283

Chapter 26 - Software Testing for Radar & Electronic Warfare Systems....................................294

Chapter 27 - New Age Requirements of Cyber Security...........................................................299

Section 6: Gain Confidence, Achieve Realism

Chapter 28 - Achieve Realism and Accuracy in EM Spectrum Operations.................................302

Appendix

Acronyms.............................................................................................................................311

Additional Resources.............................................................................................................314
Section 1

Introduction & Evolving Challenges

in Radar and EW
Chapter 1
Sophisticated Threats Raise Stakes in Electronic Warfare Conflicts
Most modern military conflicts begin beyond the sight of the naked eye. Through all domains – land, air,
sea, space, and cyberspace – military forces fight for dominance of the electromagnetic (EM) spectrum.
These forces use the EM spectrum to detect enemy forces, deceive them, or disrupt their efforts. With
different forms of electronic attack, they can weaken, disable, or even devastate their enemy’s spectrum
usage. The adaptive nature of these EM spectrum operations makes it difficult to stay a step ahead of
them. At the same time, widely available and greatly improved technology translates into more adversaries
in the electronic warfare (EW) domain. Today’s EW environment increasingly features unknown actors
producing responsive, unpredictable threats. Countermeasures must meet or exceed that pace of rapid
development to prevail in the EM spectrum.

Sophisticated Threats Raise Stakes in Electronic Warfare Conflicts 6

EW, defined as warfare in the EM spectrum, generally includes anything operating in the radio frequency
(RF). Electromagnetic operations target many aspects of the EW environment, ranging from radars and
jammers to military communications. Anything that communicates over the air is a potential target.
EW systems use the EM spectrum to support communications, sensing, and defense. Disarming these
capabilities means denying an adversary’s ability to communicate or navigate. Signal intelligence systems
also gather intelligence or find targets. In all applications, those who perform spectrum maneuvers
most quickly achieve an advantage, as modern threats work to ensure success using responsive,
unpredictable behavior.

Threats do not have to conduct a visible attack but instead cause failure in communications, coordination,
and other operations. Due to the convergence of the following trends, threats have grown increasingly in
number and sophistication:

1. Availability of technology: 10 years ago, very few players dominated this battlefield. The technological
capabilities and investments required to dominate in EW prohibited others from developing competing
EW capabilities. As commercial electronics became cheaper and more available, adversaries of all sizes
entered the EW arena. Now, even smaller adversaries potentially have a competitive threat arsenal,
making the threat environment more dangerous and unpredictable. With the barrier to entry so low,
anyone with the right skill and knowledge can secure enough equipment to be a threat.

2. Software-defined radio (SDR) systems: Originally, SDR translated to a reconfigurable radio-based

only software. Analog-to-digital conversion occurred directly at the antenna. Modern SDRs often take
more complex forms, changing their operating frequency, modulation, operating bandwidth, and network
protocol without having to change the system hardware. As speeds increase for both digital signal
processing (DSP) and analog-to-digital converters (ADC), more signal processing occurs digitally. By
leveraging such systems, military forces can more easily upgrade their threat systems. The rapid pace of
change for commercially available dual-use technologies and software-defined systems drives much of
the diversity and complexity of future threats.

3. Artificial intelligence (AI): The pace of change for commercially available dual-use technologies and
software-defined systems drives both the diversity and complexity of future threats. With the addition of
AI, those threats also learn from each conflict and are more likely to prevail in the future.

Types of EW Threats
Threats of the past were static – consistent in appearance and behavior. Today’s threats are responsive,
changing their behavior based on the scenario. If an adversary is jamming a reactive threat, for example, it
will switch frequencies or take another action to elude that jamming. Adversaries must now assume that a
threat might change and prepare to react accordingly.

Sophisticated Threats Raise Stakes in Electronic Warfare Conflicts 7

Cognitive and Adaptive
Cognitive or adaptive threats change and adapt over time. Although people use these terms
interchangeably, many levels of adaptability exist. Most of them do not come near the capabilities
of cognitive EW. Cognitive EW systems use machine learning to enter an environment with no prior
knowledge of the adversary’s capabilities and rapidly understand the scenario. By doing something that
makes the adversary’s system react, they can quickly evaluate it and develop an effective response suited
for that particular adversary’s system.

In contrast, adaptive solutions cannot rapidly grasp and respond to a new scenario. For example, an
adaptive radar senses the environment and alters its transmission characteristics accordingly, providing
a new waveform for each transmission or adjusting pulse processing. This flexibility allows an adaptive
radar to enhance its target resolution. Many adversary systems require only a simple software change to
alter waveforms, which adds to the unpredictability of waveform appearance and behavior. Military forces
struggle to isolate adaptive radar pulses from other signals, friend or foe.

Much adaptive and cognitive development is in response to anti-access / area denial (A2/AD) threats. A2
threats prevent or hinder the deployment of ally forces into the conflict zone. For example, these threats
may do something to force the adversary to engage from a less effective distance. Successful A2 threats
prevent access or even passage of ally forces.

In contrast, AD threats take actions that impede friendly operations. These threats work to limit or
eliminate an adversary’s ability to respond to its allies or other friendly forces. Examples of such
threats include attack vessels or aircraft, as well as missiles. AD threats affect the operations within
the domain. As these threats grow increasingly adaptive, their opponents must respond to them in a
much shorter time.
Impact of Machine Learning
With AI, intelligent machines work and respond much like humans to perform more complex tasks using
capabilities like signal recognition. Machine learning takes AI one step further, allowing machines to
continuously learn from data and adapt as a result. These computers learn over time at a very rapid rate.
Threats using machine learning continue to learn from every conflict and determine ways to prevail against
future countermeasures. As the computer decides how to alter behaviors, evolution occurs without the
need for human interaction. Due to the unpredictable behavior of the threat system, even the people who
implemented it cannot foretell its exact behavior.

As threat systems advance with machine learning technology, they will adapt and alter their behavior
or course of action at an increasingly rapid rate. For example, if a radar is trying to track a jet, the
adversary’s countermeasures may stop it from succeeding. Using machine learning, that radar would
repeatedly try new approaches to achieve success.

The Countermeasure Response

EW systems respond to threats with countermeasures. By using the EM spectrum or directing energy,
forces can control the spectrum to either attack or defend. An electronic countermeasure (ECM) or attack
prevents an opponent from leveraging the EM spectrum to succeed in their operations. The goal is to
weaken, neutralize, or destroy the enemy’s capabilities with attacks on people, equipment, or buildings
and locations. These attacks use EM, directed energy, or anti-radiation weapons.

An ECM is an electrical or electronic device designed to trick or deceive radar, sonar, or other detection
systems, like infrared (IR) or lasers. It may be used both offensively and defensively to deny targeting
information to an enemy. The system may make many separate targets appear to the enemy or make the
real target appear to disappear or move about randomly.

Most air forces use ECM to protect their aircraft from attack. Military ships also deploy ECM, while some
advanced tanks recently began leveraging them to fool laser/IR guided missiles. Frequently, an ECM
is coupled with stealth advances to simplify the work of the ECM systems. Offensive ECM often takes
the form of jamming. Self-protecting (defensive) ECM includes using blip enhancement and jamming of
missile terminal homers.

As threats grow more sophisticated, countermeasures must become just as sophisticated. Radars go
through a step of evolution and emerge more sophisticated. The role of countermeasures is to counteract
these developments, so they go through their evolution to again take the lead. As the two sides compete
against each other, every step increases complexity.

Sophisticated Threats Raise Stakes in Electronic Warfare Conflicts 9

Jamming Techniques
A radar works by sending out a pulse. That pulse reflects off things in the environment and bounces back.
If you send out a pulse in the air and it encounters a fighter jet, RF energy bounces off that jet, returns
to your radar receiver, and you see the return. That return is known as an echo return or a skin return of
a platform. You jam the radar with broadband noise by putting a lot of RF energy into the receiver at the
same time that skin or echo return hits the receiver. The skin return is not detectable because of the high
energy levels.

With range gate and velocity gate pull offs, the returned pulse looks like the skin return, but it changes it
slightly. It makes it look like the direction that that aircraft is flying is different than what it is. For example,
it could appear that the reflected object is moving away instead of toward you. As a result, you lose the
tracking on it.

With digital radio frequency memory (DRFM) techniques, you take RF in, digitize those signals, and create
new RF energy based on the pulse you receive. You use this very fast turnaround technology to create
multiple false targets.

Stealth technology uses a different approach. It combines the geometry of the plane with materials that
do not reflect RF energy well. When radars send out the pulse, they do not get a good skin return back.
The combination makes it challenging for radar to detect stealth units.

Countermeasures continue evolving more rapidly to stay ahead of threats. Many of these developments
focus on DRFMs and the use of more sophisticated digital signal processing (DSP). As DSP grows more
scalable and powerful, more intelligence can be added to EA receivers. To learn how to respond to new
threats, these systems also use machine learning or AI-type applications.

In the increasingly crowded spectrum, uncertainty grows over the guaranteed success of any electronic
countermeasure. To evaluate electronic attacks or electronic countermeasures, you should generate the
threat and analyze your response to it. If accurate threat detection and analysis are enabled, your systems
will respond to threats with the right countermeasures. This ECM analysis tells you if you can prevail in
your electromagnetic spectrum operations.

Sophisticated Threats Raise Stakes in Electronic Warfare Conflicts 10

Preparing for the Future
Due to the abundance of new, modern, and responsive threats, military forces vie for control of the
EM spectrum. Spectrum dominance enables them to detect, deceive, and disrupt enemy forces while
protecting their military. Military forces must constantly innovate their EW threats and countermeasures to
dominate the EM spectrum and stay in the leading position. To keep pace with the ever-changing threat
environment, military forces demand flexible, scalable solutions. A risk mitigated today may not be an
issue six months from now. As a result, military forces continue to face new threats from enemy forces.

Incomplete, disaggregated data prevents military forces from attaining or creating a clear threat
picture. They lack a methodology against which to test these threats. This issue stems from traditional
EW threat simulation systems. They use databases of known threats, which usually have associated
countermeasures. Such classified lists of known targets are no longer as effective as they quickly fall out
of date. These systems were not built to identify and isolate threats in the EM environment and determine
countermeasures on the fly.

Even when capable of processing new signals, traditional EW threat simulation systems involve a very
time-consuming process. For example, military forces collect information from the field about a type of
signal, such as frequency or pulse repetition interval (PRI). They send that information to a lab, where it is
analyzed to gather more information and develop countermeasures. Months pass before that information
is available in the system for use.

In the future, adversaries will have a more complete picture of operations. Building on the past decade’s
transformation, the next 10 to 20 years promise to deliver faster, more evolved technology developments.
Many predict that machine learning and artificial intelligence will drive powerful, continuous evolution in
EW. The EW threat environment will leverage drastic processing improvements like using multiple devices
to provide more information in less time. Sensing technologies will also play a larger role, gathering
information about the conflict zone. New coding techniques already result in increasingly complex,
interconnected, and correlated sensors.

These technology innovations will spawn knowledgeable, newly responsive threats that find novel ways to
gain power in the EM spectrum. While the technologies will continue to evolve, and new threats emerge,
one constant remains – the military force that achieves and maintains spectrum dominance will dominate
the EW domain.

Sophisticated Threats Raise Stakes in Electronic Warfare Conflicts 11

Additional Resources:
For more information on Keysight’s radar and EW solutions, visit www.keysight.com/find/AD

GREATER CONFIDENCE THROUGH EW THREAT SIMULATION AND ANALYSIS

https://www.keysight.com/see/radar-ew-system-design

Sophisticated Threats Raise Stakes in Electronic Warfare Conflicts 12

Section 2

Virtual Prototyping
Using Design Tools
Introduction
Gone are the days of simple design. Every year, designers push new limits: higher output power,
better efficiency, smaller components, and increased levels of integration.

With complex design comes new challenges. Designers spend hours setting up and running
simulations. Mountains of data wait to be measured and analyzed. Engineers need to create
workarounds to connect multiple design tools. Meanwhile, wireless standards are evolving quickly.
To keep up with the strong demands of modern technology, a new approach is essential.

The Next Generation of Electronic Design and Test

Connected, Agile Design and Test is a transformational way to approach the development of electronic
systems. It combines new software, workflows, and powerful automation tools in a way that transforms
processes and yields substantial productivity and equipment utilization improvements.

The approach moves organizations from siloed design and test steps to agile, connected workflows. The
benefits mirror those of agile software design and DevOps:

• faster device design

• translation of design parameters into test requirements
• execution and validation of test results

When coupled with automation, what results is a new development culture known as TestOps. (Read
more in The TestOps Manifesto: A Blueprint for Connected, Agile Design and Test.)
The Transformation Starts with Design
There is a tremendous opportunity to reduce time-to-market across the design and simulation phases of
the electronic product development lifecycle. Most challenges slowing the lifecycle today can be distilled
down to data movement and tool integration. Information sharing across the workflow is one of the
biggest challenges for design and test engineers. 9 out of 10 companies revealed that correlating test
result data with simulation takes months.

The reason data correlation takes so long is primarily because of the numerous tools used throughout
the development lifecycle. Over 50 percent of designers use more than 5 different tools for simulation
and design. The software tools are not integrated and require hours of coding each week to enable data
sharing. The magnitude of that integration effort is amplified by the fact that nearly every company is
devoting more resources to the maintenance of in-house tools. Designers are looking for an integrated
solution that leverages shared data to accelerate their electronic design.

The Keysight Solution

Keysight offers a collection of electronic design automation software tools that accelerates product
development by reducing the time engineers spend in the design and simulation phase. Its libraries
and customized simulators reduce setup time. The software seamlessly integrates circuit design,
EM simulation, layout capabilities, and system - level modeling, reducing time spent in importing and
exporting designs and fixing errors associated with changing tools. Improvements in data analytics allow
for faster examination and expeditious design decisions. Automation improvements reduce manual work.
This chapter will cover how to bring efficiencies into the RF and microwave design flows, shortening the
design cycle and reducing project delays, as well as, provide a series of examples on how to apply this
design flow for radar and EW systems.

KEYSIGHT PATHWAVE DESIGN AND TEST SOFTWARE PLATFORM WHAT’S NEW IN PATHWAVE SYSTEM DESIGN (SYSTEMVUE)

https://www.keysight.com/see/radar-ew-system-design
Chapter 2
Radar and EW Development Using Model-Based Engineering
This chapter describes a design methodology and flow that is well suited for use in model-
based engineering. This is an approach to engineering that uses models as an integral part of
the technical baseline that includes the requirements, analysis, design, implementation, and
verification of a capability, system, and/or product throughout the acquisition lifecycle. It is targeted
at system engineers, directors of engineering and equivalent job functions who are involved in the
development of radar and EW systems.

Radar and EW Development Using Model-Based Engineering 16

Introduction to Model-Based System Engineering (MBE)
The radar/EW lifecycle comprises of development, deployment, maintenance, and enhancement for
future requirements; each must be considered upfront for developing an effective and efficient system
and to achieve low cost of ownership. Highly complex radar and EW systems have been developed and
deployed in complex environments with great performance and superb results since World War II. Since
then, however, signal complexity has gone up by several orders of magnitude and new technological
innovations have been introduced to realize ever increasing capabilities. These two aspects alone
make the radar and EW systems’ development lifecycle very complex. In fact, without modern tools
and processes, it would be almost impossible to meet today’s delivery times and budget requirements.
Because of this, several big organizations such as Raytheon, Northrop Grumman, Lockheed Martin,
Boeing, and others—came together to evolve new methods for radar and EW systems development.
Model-Based Engineering (MBE) is one such technique.

This chapter reviews the MBE technique. It also examines in detail how a set of simulation and modeling
tools can work together to support MBE. 

MBE Basics
As per the final report1 of the MBE subcommittee of National Defense Industrial Association (NDIA), MBE
is defined as:

“An approach to engineering that uses models as an integral part of the technical baseline that includes
the requirements, analysis, design, implementation, and verification of a capability, and/or a product
throughout the acquisition cycle.”

During development, therefore, all aspects of radar acquisition have to be considered. MBE was originally
derived because of several gaps in the practice and the process of developing complex systems. While
the negatives may have necessitated MBE, the advantages have accelerated the focus and attention on
it. MBE’s main advantages include the ability to:

• quickly evaluate what’s not possible by exploring the entire project scope
• derive all downstream activities
• come up with quick proposals and Rough Order of Magnitude (ROM) estimates
• rapidly evaluate the system performance for changing requirements
• use the top-level model as the beacon for every subsystem development

With a system model, it’s possible to derive all subsystems’ requirements and partition the design margins
more meaningfully; a process generally known as top-down design. In the case of MBE, the system
model not only guides the subsystem design, but also helps define the system test, manufacturability and
cost of production, among other things. This also leads to the ability to generate a quick proposal that is
totally devoid of impossibilities and has a decent probability of success.

Radar and EW Development Using Model-Based Engineering 17

What are the impossibilities in the context of radar? Certainly, radar cannot have infinite range and a pulse
radar requires a minimum range dictated by the finite speed-of-light. A pulse Doppler radar does not
receive while transmitting. Hence, any signal that returns during transmission is not detected. This limits
the detection range and this limitation comes from the physics of the system. Similarly, we can identify
other limitations associated with such things as resolution and Doppler ambiguity. With a good master
model upfront, one can quickly examine these limitations early in the design phase.

Clearly the first, and perhaps the only, critical step in MBE is the development of the model. When
developing the ‘model’ we need to consider two aspects:

• Width and versatility. How wide should the model be? The radar model must predict the performance
under a variety of conditions, such as frequency range, complex environments, waveforms, wide
variety of threats, and various interferences. It is generally desirable to build a model as wide and
versatile as possible so that one can explore all the above conditions.
• Depth. How deep should the model be? This is an engineering and business decision. There are
tools that allow both the fidelity and accuracy of the building block models to be increased infinitely.
Doing so; however, would increase the simulation time enormously and could be prohibitive and
counterproductive from a business point of view. Hence, a good system model is one that has
enough accuracy, but that also simulates quickly. A good system model also has fidelity that can be
progressively increased on demand.

Let’s now go to the question of: What is being engineered? In the context of radar, this can be several
systems. We can view radar as a system of systems as shown in Figure 2.1.

DSP research /
engineering

Data Antenna
processing engineering
engineering

System
engineering

Mechanical RF/µW
engineering engineering

Power
engineering

Figure 2.1. Radar can be viewed as a system of systems.

For the rest of the discussion we will consider only the electronic system of radar, which includes the DSP
and RF/MW subsystems in the transmitter and receiver, and the antenna. However, the methodology
applies to all other systems each requiring a set of special tools.

Radar and EW Development Using Model-Based Engineering 18

The MBE approach can be efficiently described by a V-diagram (Figure 2.2). The V-diagram is adopted
by systems engineering and is available from various papers and text books. The specific one shown in
Figure 2.2 is from the presentation at the INCOSE workshop2.

Operational concepts Field test

Architectural specification System integration and test

Sub-system design Subsystem test

Device/circuit design Device testing

Realization

Figure 2.2. A V-diagram of an MBE approach.

Figure 2.2 shows both a top-down and bottom-up flow. The definition and decomposition flow dictates
the specification of the subsystems and their margins. It is common that the subsystems themselves are
complex and hence, they can also be described with individual V-diagrams as shown in Figure 2.3.

High-Level System Model

Realization
Realization

Realization

Realization

Realization
Realization

Realization

Figure 2.3. Individual V-diagrams of the individual subsystems in an electronic system.

Radar and EW Development Using Model-Based Engineering 19

Figure 2.3 implies that the subsystems themselves can be complex and as such need to be treated as
individual systems that should also follow the MBE approach. For example, the phased-array subsystem
in an antenna is very complex and needs to be treated with MBE. Similarly, the Transmit-Receive (TR)
module, designed as an integrated circuit, may have to be treated like a system and all the system
concepts should be applied.

Each of these subsystems must be designed using an appropriate tool. Let’s call the V-diagram one
level below a child-V, and the one above child-V, a mother-V. An important requirement of MBE is that a
mother-V be able to call, command and control child-V. It should be able to simulate child-V, keep it in
standby mode, examine the output from child-V, and change the inputs to child-V to continue with the
simulation on demand.

The tools used to design each of these systems—which by itself is a subsystem for the higher-level
system—are very likely different tools. For example, a system-level simulation is done using popular
tools like Keysight’s PathWave System Design (also known as SystemVue), while the subsystem (e.g., a
TR module) is designed using a circuit simulator, for example with PathWave Advanced Design System
(ADS). Depending on the situation, some of the subsystems could be actual hardware connected to
measuring instruments. It quickly becomes clear that for maximum effectiveness, however, all these tools
should be inter-operable. In other words, a higher-level system simulating tool should be able to call a
subsystem simulating tool and put it in a command-and-control mode.

Once the above requirements are satisfied, one can build the system model to simulate the end-to-end
performance of the system. Let’s see how such a system model can be created for a radar using a
set of tools.

On close observation of the V-diagram in Figure 2.2, notice that the left side of the V pertains to top-
down design, while the right pertains to integration and verification. Verification is usually done with actual
hardware, meaning that it is not performed until the actual hardware is available. With modern tools, a
top-level model can be built that allows verification to be done in simulation. To highlight the difficulty,
imagine having to co-simulate RF and DSP circuits. However, very few tools in the industry support this
kind of co-simulation. With an improved co-simulation capability though, both sides of the V-diagram
(top-down design and bottom-up verification) can be done entirely in simulation. Final verification is then
done with actual hardware. Performing the entire V in simulation eliminates major risk factors early in the
development cycle.

Radar and EW Development Using Model-Based Engineering 20

Building a System Model Using SystemVue, MATLAB, and STK
To better understand how to build a high-level system model, consider the example of a mono-static
pulse Doppler radar, including the radar platform and target. We will show that this model can be easily
expanded to bi-static and multi-static radar systems, as well as multiple targets. We will also show how
clutter, jammer, interferer, and noise can be added to the target return. The radar and the environment
being built will cater to a typical operational scenario depicted in Figure 2.4.

Various types of clutter, interference, and jamming

Chaff
Jamming

Rain

Sea/ocean

Figure 2.4. This image depicts the typical operation of a mono-static pulse Doppler radar.

From Figure 2.4, it’s apparent that the radar is on a moving platform, a ship, and that there are three
targets, one jammer and the clutter caused by sea, rain and chaff. The communication signals from
the city environment may act as interferers. Also, the radar could be treated as bi-static with the radar
receiver on the reconnaissance aircraft. It is important to note that pictures like that shown in Figure 2.4
have to be made visible throughout the organization engaged in MBE. Everyone then will be able to check
the performance of their individual subsystems through the system model.

Radar and EW Development Using Model-Based Engineering 21

Target return, jamming, clutter and noise: The scheme for modeling the target reflection, jamming signal,
clutter, and noise is shown in Figure 2.5.

Figure 2.5. Shown here is the scheme for modeling the target reflection, jamming signal, clutter, and noise.

The inputs to the target return model—the target position, target RCS, transmitting antenna position,
receiving antenna position, and waveform from the source—come from the antenna layer. The output
from this model is computed using the radar equation.

The Jammer model is versatile and allows both cover and deceptive jamming to be modelled. There
are four types of cover jamming: barrage, spot, multi-spot, and swept spot. For barrage jamming, the
model generates the Additive White Gaussian Noise (AWGN) with mean and standard deviation values
determined by the parameters Mean and Stdev in the full bandwidth. For spot jamming, the model
generates the band limited AWGN noise. The normalized bandwidth is determined by the parameter
Bandwidth. Also, the FilterTapsLength is used to set up the maximum length of low-pass filter. For
multi-spot jamming, the model generates band limited AWGN noise in the sub-bands determined by the
parameter MultiSpotBand, which can be used to set up the normalized start and cut-off frequencies of
the passbands. For swept-spot jamming, the band-limited AWGN noise sweeps in the scope of the full
band, with the sweep rate decided by the parameter SweepFreqStep.

Radar and EW Development Using Model-Based Engineering 22

The clutter model is built from the probability of distribution (PDF)
and power spectral density (PSD) functions. It is designed to
NOTE. The idea here is
not to absorb all of the
generate the correlated coherent and non-coherent clutter. The
details of the model, but to
model, which supports Gaussian, LogNormal, Weibull, and K clutter emphasize that in MBE the
amplitude distribution, is a computationally intensive model. When models have to be made
used in the workspace, it causes the simulation speed to slow. as wide as possible.

The clutter model is different from noise in two major ways. First,
the power spectrum of clutter is not white and is the result of echo.
Second, the model supports the Gaussian, all-pole and Cauchy
power spectrum model.

Figure 2.6. Measured or simulated clutter files can be used for a given terrain.

The measured clutter files, in terms of I and Q, are read from the
external files into the simulation. NOTE. While the model
is versatile, it can take
After the addition of noise, the return signal goes into the radar a long time to simulate
receiving antenna. The receiving antenna is very similar to the and this is contrary to
transmitting antenna in that it takes input from the antenna layer the MBE system model
and forms a beam in the direction of the target. The output of the philosophy. Hence, a better
receiving antenna is passed through an RF down-converter into way is to use measured or
simulated clutter files for
the baseband processor. Adaptive digital beamforming techniques
the given terrain. One such
are used to combine the outputs of various phased-array antenna
implementation is shown
elements to form a signal input for the pulse compression block. in Figure 2.6.

Radar and EW Development Using Model-Based Engineering 23

Increasing Subsystem Model Fidelity
While the system model is useful for the quick top-level simulation, it should also support verification
of the subsystems designed with higher fidelity models. As an example, the power amplifier (PA) in the
transmitter might be designed by a circuit specialist and it may be desirable to see how the actual PA
circuit performs in the top-level model (Figure 2.7).

Figure 2.7. To see how the actual LNA circuit in the receiver performs, ADS can be used to simulate it. SystemVue
enables co-simulation with ADS.

A circuit simulator like PathWave Advanced Design System (ADS) can simulate actual circuits in the
design. SystemVue can co-simulate with ADS. It is also possible to scale the fidelity of the circuit model.
To do this, we extract the X-parameters* of the circuit and place it as a model for the amplifier in the
system simulator.

An example can also be provided for baseband models. We simply substitute HDL code for a building
block that considers the finite precision accounting for the quantization effects. Another example is to
read in the measured antenna patterns for the antenna elements rather than assuming an analytical
antenna factor. The choice of varying fidelity for the building block models is a key requirement of MBE.

* “X-parameters” is a registered trademark of Keysight Technologies. The X-parameter format and underlying equations are open and
documented. For more information click here. (https://www.keysight.com/in/en/lib/resources/miscellaneous/x-parameters-open-
documentation-trademark-usage-and-partnerships.html)

Radar and EW Development Using Model-Based Engineering 24

Bringing a More Dynamic Environment into the System Model
The software platform can also co-simulate with the STK simulation tool from Analytic Graphic
Incorporated (AGI), an Ansys company. STK is a physics-based software geometry engine that accurately
displays and analyzes land, sea, air, and space assets in real or simulated time.

First, users model the time-dynamic position and orientation of vehicles. Given these dynamic positions
and orientations, users can model the characteristics and pointing of sensors, communications and other
payloads aboard the asset. STK then determines the spatial relationships (e.g., line-of-sight) between
the assets of interest and all the objects under consideration. These relationships can also be modeled
across multi-hop links or over regions of interest. STK assesses the quality of these relationships through
a wide array of constraining conditions (e.g., payload capability, unique user algorithms, etc.), while also
incorporating environmental effects such as terrain, lighting and weather conditions on sensor visibility or
communication link quality.

When SystemVue is co-simulating with STK, the inputs are brought from STK directly into SystemVue’s
signal layer. For more details on this process, please refer to reference 3.

Any tool utilized to build a system using the model-based engineering technique requires three key
characteristics. It must be able to create the wide and versatile models needed to build a system model.
It must support scalable fidelity for the models. And, it must have the ability for an external program to
command and control the system model.

References
1. Final report of the Model Based Engineering (MBE) subcommittee, NDIA systems engineering
division, M&S committee, February 10, 2011.
2. Ron Williams, INCOSE Model Based Systems Engineering, Integration and verification scenario,
Raytheon, INCOSE IW12 MBSE Workshop, January 21-22, 2012.
3. Virtual flight testing using SystemVue and STK, 5991-1254EN, Keysight Technologies.

RADAR CASCADE ANALYSIS WITH PATHWAVE SYSTEM DESIGN

COURTESY OF MICROWAVE JOURNAL
https://www.keysight.com/see/radar-ew-system-design

Radar and EW Development Using Model-Based Engineering 25

Chapter 3
Radar and EW System Design and Interference Analysis
This chapter outlines an example of PathWave System Design (SystemVue) software for
performing radar and EW system design and jammer/interferer analysis. Some of the key areas
to be discussed include how to implement a radar chirp waveform, design an RF chain for the
transmitter and receiver, and perform pulse-compression analysis using fast Fourier transform
(FFT) based convolution. Finally, the radar system is tested in the presence of unwanted
interference and jamming signals to study the impact of such unwanted impairments on
radar performance.
Custom Signal Generation
LFM chirp for radar system design
SystemVue offers a flexible platform to create custom signals. In the example shown in Figure 3.1,
SystemVue floating-point Data Flow components were used to model the linear frequency modulation
(LFM) chirp source. The integrator on the left increments time until the value for the pulse period is
achieved, causing it to reset and start over. The values for u (μ) and wc (ωc) are computed as shown
in Figure 3.1.

pulse pulse 11
Gain=31.42e+6 [wc] A1
u=2*pi*(FM_High.FM_Low)/PW;
Sqr wc=2*pi*FM_Low;
7 10
Math Gain=31.42e+12[0.5*u]
FunctionType=Sqr

(3.1a)

(3.1b)

Figure 3.1. Custom signal generation using SystemVue DSP library blocks

Radar and EW System Design and Interference Analysis 27

Custom signal generation using MATLAB script
SystemVue offers built-in MATLAB script language to be used throughout the program. In Figure 3.2, the
LFM chirp source is defined in a MATLAB_Script component.

Figure 3.2. Custom signal generation using MATH Language in SystemVue

Custom signal generation using third-party tools

SystemVue offers direct links with C++, HDL, and MATLAB. Any custom signals written using these
languages can be easily brought into SystemVue as shown in Figures 3.3 and 3.4. Co-simulation with
MATLAB allows the user to incorporate pre-existing m-code files.

(3.3a. MATLAB co-simulation link (3.3b)

Figure 3.3. Linking MATLAB script with SystemVue

Radar and EW System Design and Interference Analysis 28

Figure 3.4. Custom waveform code in C++

LFM Chirp IF Generation

In this chapter, Math Language is used to generate the LFM waveforms used for the radar system
design (Figure 3.5).

Note that the M-Code component does not provide SystemVue with the sample rate information that
its built-in models do. It is recommended, therefore, that a sample rate component be added after
a M-Code component that is used as a source to implicitly define the sample rate (Figure 3.5). The
CxToEnvelope component defines the complex waveform as an RF Envelope waveform where I&Q,
plus time and carrier frequency are defined. The spectral result, centered around 500 MHz, is shown
in Figure 3.5.

Figure 3.5a Transmitter IF Spectrum, centered around 500MHz

Radar and EW System Design and Interference Analysis 29

Figure 3.5b

Figure 3.5. LFM chirp signal generation and IF spectrum. Note that the Sample_Rate component declares the time-
step for the un-timed complex numeric data emerging from the MATLAB Script source.

Transmitter RF Design
The RF chain of the radar transmitter was designed using the Data Flow RF block library in SystemVue,
which offers users a variety of RF models to implement the RF section of the system (Figures 3.6a and
3.6b). Real-world impairments such as nonlinearities, LO phase noise, and mixer leakage products can
also be incorporated into the simulation.

Figure 3.6a. Transmitter RF spectrum

Radar and EW System Design and Interference Analysis 30

Figure 3.6b. RF section of radar transmitter using the RF block library in SystemVue

Radar Propagation Loss Modeling

The radar propagation path modeling was performed using SystemVue library blocks to implement
standard math equations for propagation delay and free-space propagation loss (Figure 3.7a and 3.7b).

Figure 3.7a. Radar signal as seen at receiver input

Figure 3.7b. Implementing radar propagation loss and propagation delay models

Radar and EW System Design and Interference Analysis 31

Radar Receiver Design
The radar front-end (FE) receiver was designed using RF blocks. Various budget analyses were performed
to optimize the performance as shown in Figure 3.8a to Figure 3.8e.

Figure 3.8a. Radar front-end receiver design using RF block library in SystemVue

Figure 3.8b. Noise-figure budget analysis of receiver front end

Figure 3.8c. Channel-power budget of receiver front end

Radar and EW System Design and Interference Analysis 32

Figure 3.8d. Cascaded-gain budget of receiver front end

Figure 3.8e. Receiver output IF Spectrum

Radar and EW System Design and Interference Analysis 33

Receiver Signal Processing
The receiver IF was filtered, down converted, and down sampled for analog-to-digital conversion.
The post processing algorithm was designed using SystemVue DSP library blocks as shown in
Figure 3.9a to Figure 3.9e.

CIC Filter
0=Deactive (Short)
Receiver IF 1=Activated

From FFT Based

Chirp
Source Convolution
Solver
Figure 3.9a. Top-level view of receiver IF post processing

FFT Based Convolution

Take Congujate
of Reference Signal
Reference In
Divider.

Reference/Test Out=conj(Ref)*Received

Inverse FFT

Integrate FFT
magnitude values 1/Test so that as
Test increases
VGA gain
decreases
Note: The two
input port
locations are
reversed from
the symbol

Signal Levels are

normalized at this
Received Signal point

Figure 3.9b. FFT-based convolution algorithm (pulse compression) using SystemVue DSP library blocks

Radar and EW System Design and Interference Analysis 34

Figure 3.9c. Time-domain waveform at ADC input

Figure 3.9d. Time-domain reconstructed data at ADC output

Figure 3.9e. Compressed pulse after post processing

Radar and EW System Design and Interference Analysis 35

Jammer/Interference Analysis
Jammer/interference analysis can be performed on the radar system design as shown in Figure 3.10a to
Figure 3.10d. Note that jammers/interferers can be varied to have the different levels of amplitude and
frequency needed to perform radar receiver fidelity analysis and what-if analysis.

Figure 3.10a. Creating and combining interferer source with radar waveform

Figure 3.10b. Composite spectrum at radar receiver input

Radar and EW System Design and Interference Analysis 36

Figure 3.10c. Various waveforms in the presence of jammers/interferers

Figure 3.10d. Sweeping interference power to see its effect on compressed-pulse peak detection

Radar and EW System Design and Interference Analysis 37

Integrating Software and Instruments for Advanced System-Level Design
While it is always possible to design and test complex defense systems separately, ideally the
engineer would like these two domains to be interlinked to enable a fully integrated approach to
system development and testing. This can be accomplished using the native instruments link option
in SystemVue. This option allows both the designer and test engineer to link a variety of test and
measurement equipment with SystemVue to achieve the following objectives:

• Create and download arbitrary signal to instruments

• Create near real-life signals for system integration testing in the lab environment
• Capture signals from instruments and design remaining blocks by taking real device distortions
into account

Integrated test setup

A fully integrated approach to system development and testing can be used to address the case study
presented in this chapter. The test setup, as shown in Figure 3.11a, requires:

• SystemVue software installed on a PC

• A Keysight vector signal generator (VSG), in this case, the N5182A MXG
• A Keysight PXA Signal Analyzer, in this case, the N9030A PXA

In this case study, the LFM chirp radar waveform is used. It is designed using SystemVue software with
different specifications and downloaded onto the VSG (Figures 3.11b and 3.11c). SystemVue allows data
to be downloaded from any node in the design. In this case, the node selected is the one where there is
a combined spectrum of main radar return coming to the receiver input and where interfering signals can
be added to perform receiver signal processing fidelity test (Figure 3.11d).

Figure 3.11a. Custom signal download to a VSG

Radar and EW System Design and Interference Analysis 38

 Insert Signal Downloader
Component
Transmitter and RF Path
Chirp Rader Waveform
Polymporphic Receiver
Model - Change
representation between Data
Flow or RF model
implementation.

CIC Filter
Interference Sources 0=Deactive (Short)
1=Activated

From
Chirp
FFT Based
Source Convolution
Solver

Figure 3.11b. SystemVue workspace with signal downloader component to download custom waveform to VSG

Figure 3.11c. Ideal LFM chirp waveform being analyzed using spectrum analyzer and its vector signal analysis using
89600 VSA software running on the N9030A PXA

Radar and EW System Design and Interference Analysis 39

Figure 3.11d. Composite LFM radar with interferer signal and interference power sweep to understand receiver
behavior under different conditions

Radar and EW System Design and Interference Analysis 40

Chapter 4
Simulation and Verification of Pulse Doppler Radar Systems
Modern radar systems that operate in environments with strong clutter, noise, and jamming
require advanced digital signal processing techniques. Direct analysis techniques often fail when
designing such complex systems. Although simulation is often used, most simulation tools do
not have enough models and integration capability to handle modern radar systems.

This section proposes a solution to this dilemma, a system-design methodology that uses
PathWave System Design (SystemVue). Examples will be used to illustrate how advanced pulse
Doppler (PD) surveillance radar with moving target detection (MTD) and a constant false alarm
rate (CFAR) processor can be designed. To ensure the design works properly, the platform can
be connected to instrumentation for system test and verification. This allows users to reduce
their system development time and cost, while also decreasing their chances of unexpected
system failures late in the system development process.

Simulation and Verification of Pulse Doppler Radar Systems 41

System Design Challenges
Advanced radar systems are very complex, necessitating sophisticated signal processing algorithms.
Effective algorithm creation requires both a platform for simulation and verification. Models for signal
generation, transmission, antennas, T/R switching, clutter, noise, jamming, receiving, signal processing,
and measurements are also needed to create advanced algorithms. Most simulation tools do not have
enough models and the integration capability needed to design such complex algorithms.

Signal Processing Algorithm Creation

The system-design methodology proposed below uses SystemVue. It concentrates on algorithm design
for PD systems. The development of MTD and CFAR processors in a time-efficient manner are used as
examples to better understand this methodology.

To begin, consider the tasks undertaken by the radar DSP algorithm designer, which typically breaks
down into the following two stages:

Stage 1. Design the algorithm in software and verify it using a simulation tool.
To accomplish this task, the designer needs:

• A user-friendly algorithm modeling environment to easily create and debug the algorithm during
development. The environment should support multiple languages, such as m-code, C++, and HDL.
• Signal sources and measurements to verify the algorithm during verification, once it is created.
Unfortunately, it is not a simple task for the DSP algorithm designer to create radar signal sources
with radar cross section (RCS), clutters, noise, and jamming. The designer might also encounter
difficulty creating measurements for the algorithm. Consequently, a tool that provides radar sources
and measurements is desired.

Stage 2. Implement and test the algorithm in hardware.

After the algorithm testing and verification are finished, it needs to be implemented in hardware. To
accomplish this task, a hardware test platform must be created.

To address these needs when creating the MTD and CFAR processor, a platform that provides the
following functionality is required:

• An interface to a vector signal generator to generate test signals. The vector signal generator
provides radar signal-generation test sources and models for RCS parameters, Clutter, Jamming,
Doppler, and Frequency offset.
• An interface to a vector signal analyzer to verify the implemented hardware as compared to the
original algorithm. The platform must support a wide range of measurements including waveforms,
spectra, detection rate, and false alarm rate (FAR). Also, it must provide an estimation of target
distance, speed, and angles for the detected target.

Simulation and Verification of Pulse Doppler Radar Systems 42

PathWave System Design (SystemVue) as a Platform for Simulation
The top-level system platform structure is shown in Figure 4.1. From the block diagram, the main models
include signal source, transmitter, antenna, T/R switch, RCS, clutter, jammer, receiver, signal processors,
and measurements. Sub libraries are also listed. The major task is how the designer creates their
algorithm using SystemVue.

Platform for simulation

Tx Waveform T/R Component Antenna Library Radar Environment Radar Signal

Library Library Processing Library
– Radar Pulse – DAC – Antenna Models – Target – Digital Pulse Compression
Generator – PA – T/R Antenna Array – Clutter – Moving Target Indication (MTI)
– Coherent Signal – Up Converter – Antenna Propaga- – Jammer – Moving Target Detection (MTD)
Generator for Single – Filter tion – Interference – Constant False Alarm Rate (CFAR)
or Multiple Channel – DDS – Digital Beamformer
– LNA – Space-Time Adaptive Processing
– Down Converter (STAP)
– ADC

Figure 4.1. SystemVue as a platform for simulation

SystemVue as a Platform for Test

SystemVue can be used as a test platform for verification of the integrated system at each stage of
development. This is done by connecting algorithms, instruments, and test hardware together based on
the created algorithm.

In SystemVue, an interface model (sink) allows for direct connections with various signal generators as
shown in Figure 4.2. This allows software data to be downloaded to instruments for hardware test data.
Figure 4.3 shows the details for using the platform to connect to instruments and test hardware based on
the created algorithm.

Simulation and Verification of Pulse Doppler Radar Systems 43

Figure 4.2. SystemVue as a platform for test

Platform – Generate test signals

Figure 4.3. Generate test signals using SystemVue

Simulation and Verification of Pulse Doppler Radar Systems 44

 Platform – Expand measurements

Figure 4.4. Expanding measurement capabilities using SystemVue

SystemVue can also connect to signal analyzers, logic analyzers or scopes to provide additional
measurements and expand instrument capability according to the user’s needs. As an example,
Figure 4.4, shows the link between SystemVue and a signal analyzer. Here, test signals from the
signal generator are sent to the device-under-test (DUT). The signal analyzer captures the DUT
output waveforms and sends them to SystemVue, using the Vector Signal Analyzer (VSA) link model.
In SystemVue, the waveform can be further processed using the radar signal processing function.
SystemVue also provides additional measurements such as Doppler frequency, detection probability, and
false alarm probability.

Principles of the PD Radar System

To better understand the proposed system design methodology, consider the design of PD radar
algorithms. PD are extremely valuable for finding small moving targets hidden by heavily cluttered
environments and are used in both military and commercial applications. Unlike continuous waveform
(CW) radar, PD radar has the ability to detect angle, distance, and velocity. Typical examples include
weather, low-flying aircraft, and anti-ship missiles.

Simulation and Verification of Pulse Doppler Radar Systems 45

For this discussion, we focus on airborne PD radar that is trying to detect a moving target near the
ground or the sea. In this case, the moving target returned signal is much weaker than the clutter
from the ground or sea. As shown in Figure 4.5, the target signal is hidden by a heavily cluttered
environment. Consequently, it is almost impossible to detect the target in the time domain using regular
radar processing.

Figure 4.5. Target signal hidden by heavily cluttered environment

The radar transmission signal and returned signal [1] can be expressed as:

S(t) = A(t)Cos(2πj ct)

S(t -1) = A(t) Cos (2π ( fc + fd) t – 1 ) + Nc(t) + Nn(t) + Nj(t)

where fd is the Doppler frequency, 1 is the delay, and Nc, Nn and Nj refer to clutter, noise and jamming,
respectively. To detect target, speed, and distance, fd and 1 must be estimated.

Because the small moving targets are hidden by the heavily cluttered environments, they cannot be
detected in the time domain. Instead, the signal must be detected in the frequency domain using Doppler
frequency analysis. To do this, return data must be collected and processed using two-dimensional (2D)
signal analysis for both target, speed, and distance. 2D signal processing is required for moving target
indicator (MTI) and MTD. CFAR processing is needed for auto-detection in PD signal processing. Without
CFAR, auto-detection will likely fail.

Simulation and Verification of Pulse Doppler Radar Systems 46

 Sampling Interval
= PRI

Sampling Interval

Range bins
= 1/BW

Doppler Frequency Extract

Figure 4.6. 2D signal processing for moving target detection

The first step in PD signal processor development is to design a data bank to store received timed
signals, as shown in Figure 4.6. The received data is entered into the data bank point-by-point from one
column to another until the bank is full.

Taking a closer look at each column, the time interval for each data point is 1/bandwidth. Each data point
is a return signal from different distances. The sampling interval between each column is the pulse signal
repetition interval. All data points in the same row are returned from the same distance with different
timing. Doppler frequency can be extracted from data in the rows. Either a filter bank or a group of fast
Fourier transform (FFT) operations can be used for all data points in the data bank. In software design,
a group of FFTs is always used. Once the Doppler frequency is detected, the location of the row can be
mapped to the return target distance from the range bins. Then, the distance can be detected.

Simulation and Verification of Pulse Doppler Radar Systems 47

PD Radar System Structure
Figure 4.7 depicts the top-level structure of a PD radar system in SystemVue. Since we are interested
in creating an algorithm for PD signal processing, this discussion focuses on PD signal processing,
including MTD and CFAR. For other blocks, just a brief introduction is provided.

PD Radar System Structure

Figure 4.7. PD radar structure

Signal sources
For this example, signal sources include:

• Pulse signal generator

• Linear FM pulse signal generator
• Nonlinear FM signal generator
• Polyphase code generator

Target return model

The target return model includes:

• RCS, Doppler effect, delay, and attenuation

• Fluctuant RCS types: Swirling 0, I, II, III, and IV linear FM pulse signal generator

Sr (t – tr) = k A (t – tr) cos [ 2 π ( f0 ± fd ) t – 4π R0 / l +r ] • U (t – tr)

where R0 is the distance between radar and the target, and tr is the path delay, so tr = 2R/c.

Simulation and Verification of Pulse Doppler Radar Systems 48

Clutter models
Since this is the system-level simulation, behavioral models can be used to describe the functionality.
We focus on the probability distribution functions (PDFs) and power spectrum densities (PSDs) that
are acceptable to model the system performance, but the physical-level model is not provided here. If
required by the user, this service can be provided. SystemVue offers a choice of four PDFs and three
PSDs, which include:

PDF PSD
• Rayleigh • Gaussian
• Log-Normal • Cauchy
• Weibull • All Pole
• K–Distribution

The user can also define any distribution using SystemVue’s built-in MATLAB Script capabilities.

Radar RF transmitters and receivers

Here we provide the behavioral model’s structure. If desired, the SpectraSys RF link enables the user to
go down to the circuit level. Another way to model complex frequency-dependent behavior is to import
S-parameters using the SData model. An example of using the SData model is shown in Figure 4.11.

As shown in Figure 4.8, the RF transmitter features local oscillators, which can include phase noise,
modulators and mixers with non-ideal behavior, and amplifiers which can include complex nonlinear
behaviors and filters. Figure 4.9 depicts the RF receiver’s oscillators, demodulator, amplifiers, and filters.

Figure 4.8. RF transmitter

Simulation and Verification of Pulse Doppler Radar Systems 49

Figure 4.9. RF receiver

Digital up/down converter (DUC/DDC) for digital IF

The two models depicted in Figure 4.10 are very useful for creating new DSP models with certain
algorithms.

Figure 4.10. Digital up/down converters

Simulation and Verification of Pulse Doppler Radar Systems 50

Antenna models
The antenna model structure shown in Figure 4.11 was created using a SystemVue SData model. If
the user knows the antenna’s Gain as a function of Deflection Angle, Modeled with S-parameters, the
simulation model can be easily structured. Figure 4.11 also shows the antenna measurements.

Figure 4.11. Antenna structure

Pulse compression
Figure 4.12 depicts the PD processing pulse compression.

Digital Compressed
DFT IDFT Digital
2D I/Q Matrix
2D I/Q Matrix

Reference DFT with Spectrum

Waveform zero padding Windowing

Figure 4.12. Pulse compression in frequency domain for PD processing

Simulation and Verification of Pulse Doppler Radar Systems 51

Moving target indicator
The basic idea behind the MTI is to filter the clutter at or very near DC, while keeping the other spectrum
region flat. As shown in Figure 4.13, a three-pulse (double, second-order) canceller can be formed by
cascading two first-order sections using a transfer function.

H(z) = 1 – 2z–1 + z–2

+ +
x[n] ∑ ∑ y[n]

z–1 z–1

Figure 4.13. Moving target indicator

Moving target detector

The MTD is a key processor for PD radar [2]. A bank of Doppler filters or FFT operators cover all possible
expected target Doppler shifts (Figure 4.14). The input data is collected in a repetition period by using a
data bank. Data points within the same range are then correlated and processed until all data in the data
bank is processed. There are two ways to administer the 2D signal processing, either using a filter bank
or a group of FFT. In the example, a group of FFTs is used to operate on each row in the data bank to
detect fd. Delay is then detected by looking at the detected data point location for range bins.

Delay Estimation for the

Range bin
Range bins

Filtering Bank Or
spectra Estima-
tion for Doppler
Freq

For Doppler Frequency

Figure 4.14. Signal processing in MTD

Simulation and Verification of Pulse Doppler Radar Systems 52

Once the algorithm is understood, the code can be derived using C++ or MATLAB Script. During its
development, the code can be easily debugged for either C++ or users can modify the code or insert
their own in the code window to implement their own MTD algorithm.

CFAR Processor
Since modern radar requires auto-detection, PD radar must use CFAR to control the false alarm rate.
Otherwise, the radar won’t work. The CFAR can be done in time, frequency domains, or both. Instead
of the fixed detection threshold, the averaging amplitude value of reference cells is used as the threshold
to prevent false alarms from happening too frequently. This CFAR system is called a cell averaging (CA)
CFAR system.

In the PD radar example in this chapter, CFAR was done in the frequency domain. Cell averaging CFAR
was used.

PD Radar System – CA CFAR

–Test cell: the detected cell

–Guard or gap cell: value not to be included in the

interference estimate due to possible target contamination
–Reference cell: values assumed to be interference only,
thus used to estimate interference parameters

Figure 4.15. CA CFAR processor code implementation

Once the algorithm is understood, the code can be derived using C++ or MATLAB Script. During its
development, the code can be easily debugged for either C++ or MATLAB Script as shown in
Figure 4.15. Users can modify the code or insert their own in the code window to implement their own
CFAR algorithm.

Simulation and Verification of Pulse Doppler Radar Systems 53

Measurements
Models have been implemented to do the following measurements:

• Basic Measurements
- Waveform
- Spectrum
- Signal noise ratio
• Advanced Measurements
- Estimation of distances and speeds
- Detection probability, Pd = number of successful detection/total number of tests
- False alarm probability, Pf = number of false errors/total number of tests
- Importance sampling will be implemented to speed up the Pf simulation [3]

If the user wants more, custom measurement models can be created using a combination of
existing models.

Simulation of PD Radar System

The PD design shown in Figure 4.7 is simulated. All key parameters can be set at the Parameter table
defined for the PD simulation system as shown in Figure 4.16. The user can very easily edit, add or delete
any parameter.

Figure 4.16. Simulation setup table

Simulation and Verification of Pulse Doppler Radar Systems 54

Simulation results
Before PD processing, at the receiver input, the target signal cannot be recognized, and the target is
hidden by a strong clutter environment (Figure 4.17). However, after the 2-D PD processing, the target is
detected. After PD and CFAR, clean target detection is achieved.

Figure 4.17. PD detection of a moving target

In Figure 4.18, the detection probability of the system is displayed. The detection rate is obtained by
running several tests and using the Pd definition to obtain it. The target distance and speed are estimated
using the detected Doppler frequency and the detected range bin location.

Simulation and Verification of Pulse Doppler Radar Systems 55

Figure 4.18. System detection rate and estimated target distance and speed

Algorithms are critical for high-performing advanced radar systems. A unified approach to radar system
design that relies on SystemVue now offers designers a viable means of creating effective algorithms.
It provides a user-friendly environment for algorithm development, while also integrating software and
hardware to verify the algorithms. SystemVue can even be used to develop algorithms for digital array
radar plus spacetime adaptive processing (STAP) and multiple-input multiple-output (MIMO) radar.
Moreover, it allows the system development team to quickly and easily try new and innovative ideas and
to evaluate the effects of jamming and interference sources on radar performance.

References
1. I. Skolnik, Radar Handbook, 2nd ed. McGraw-Hill Inc., 1990.
2. D. Curtis Schleher, MTI and Pulse Doppler Radar, Artech House, Inc., 1991.
3. Dingqing Lu and Kong Yao, Importance Sampling Simulation Techniques
Applied to Estimating False Alarm Probabilities, Proc. IEEE ISCAS, 1989, pp. 598-601.

Simulation and Verification of Pulse Doppler Radar Systems 56

Chapter 5
Applying Ultra-Wideband Technology in Radar and EW Systems
Integrating Design with Ultra-Wideband Test for Flexible Radar Verification
Ultra-wideband (UWB) radar has become increasingly popular in both commercial and defense industries.
UWB radars (whether impulse, LFM, noise, or OFDM-based) are defined as having a bandwidth of greater
than 0.5 GHz, or more than 20% of their center frequency, and are regulated by FCC rules that allow
UWB technology to coexist with existing radio services without causing interference. They offer several
advantages including high accuracy for target detection, good precision for penetrating radars, and low
cost for combining radar and communication systems. UWB radars can pass through walls and other
obstacles for geolocation/positioning and can support multipath immunity and frequency diversity with
minimal hardware modifications.

Modern UWB radar systems often operate in unpredictable environments, with interference, jamming,
and other “real world” performance limitations. Therefore, during system development, it is critical for
engineers to understand how their actual hardware will perform in these environments.

Problem
Effective radar system design requires comprehensive system validation, a time-consuming and
expensive process often necessitating costly facilities and complex measurement systems. Radar
algorithms, such as target recognition and countermeasures, need to be validated early enough to
change the signal processing hardware design. Hardware receivers must also be tested with realistic
threats and jamming scenarios. Together, these often require outdoor ranges, chambers, and real-time
hardware simulators costing tens of thousands of dollars per hour.

Unlike communication system designers, UWB radar system designers face several unique challenges,
beyond sheer bandwidth. Impulse radar signals, for example, can change shape during propagation
(e.g., non-sinusoidal waveforms), while for noise-like radars, figuring out how to model the noise in the
waveform can be challenging. With Linear FM systems, generating UWB signals with Doppler frequency
offsets, target echoes, and clutter to perform receiver verification can be challenging. As a result,
designing and testing UWB radar systems requires a variety of signal sources, target environment setups,
and measurements. Carefully designed and optimized waveforms are essential to ensure excellent real-
world performance.

Applying Ultra-Wideband Technology in Radar and EW Systems 57

Solution
To successfully develop UWB radar systems, today’s system engineers require a more flexible, lower-
cost means of validation with stimulus/response equipment that is specifically geared toward UWB. That
R&D test bed starts with Electronic Design Automation (EDA) software to model a working reference
design. The reference design is used to generate test vectors, as well as process received signals that are
captured from live measurements and organize a “model-based design flow.” The test bed also includes
a UWB signal generator with a wideband arbitrary waveform generator (AWG) to render simulated signals,
including realistic threats and jamming scenarios, for testing UWB hardware receivers. Finally, the UWB
test bed should include a wide-bandwidth oscilloscope for waveform capture.

An additional role of the EDA software is to surround the raw radar design and test equipment with the
environmental, baseband, and RF modeling required to close a round-trip signal processing loop to
perform early simulation-based verification. As hardware becomes available, the software continues to
connect directly into the physical hardware measurement. By leveraging the design tools into verification,
a consistent approach is maintained throughout the research and development process saving time,
promoting re-use, and making optimum use of the capital equipment assets.

Keysight provides an example of just such a test system. As shown in Figure 5.1, the system starts
with the SystemVue simulation and modeling environment, which is used with a wideband AWG (upper
left), the vector signal generator (lower right) and the 32 GHz oscilloscope (lower left). Together, these
components allow engineers to carefully design and optimize the UWB signals that are so critical to the
design, verification, and test of UWB radar systems.

The test bed shown in Figure 5.1 supports investigations of UWB architectures, as well as a direct
connection to test equipment for verification. It can be used to model, encode, and download UWB test
signals and also post-process received signals. The wide-bandwidth 90000 X-Series oscilloscope allows
RF engineers to measure and analyze UWB radar transmitter outputs using up to 32 GHz of true analog
bandwidth, without the need for external down-conversion. This direct approach reduces hardware
calibration, system impairments, and measurement system complexity and uncertainty.

Keysight’s E8267D PSG microwave vector signal generator features wideband baseband IQ inputs. When
combined with a wideband AWG, such as the 81180A, M9330A, or the new M8190A, the PSG provides
the flexibility necessary to create microwave and millimeter-wave signals for UWB radar scenarios, as well
as component validation.

With this test system, SystemVue generates and downloads different UWB radar test vectors to the
wideband AWG to create the necessary baseband signals. The output differential IQ signals of the AWG
are then modulated by the PSG to create an X, Ku,or Ka band test signal, to be used directly as an input
to a device under test (DUT) for the radar component test. Next, the output of the DUT is captured using
the Infiniium 90000 X-Series oscilloscope where radar measurements can be made (Figure 5.2). Signals
can be analyzed inside the Infiniium oscilloscope using the Oscilloscope Signal Analyzer (OSA) or Vector
Signal Analysis (VSA) software. For further analysis and signal processing, measured signals up to 32 GHz

Applying Ultra-Wideband Technology in Radar and EW Systems 58

wide can be brought back to SystemVue with the help of the 89600B VSA software, to close a unique
“round trip” signal processing loop (Figure 5.2).

Because of the versatility of this UWB test bed, it can be used for both validation and troubleshooting of
UWB transmitters and UWB receivers.

Figure 5.1. With SystemVue integrating this UWB test platform with a working radar reference design, engineers
can precisely generate and measure UWB waveforms for any point in a UWB system architecture, and perform
closed-loop stimulus/response measurements that reduce the need for expensive ranges, chambers, and hardware
simulators in early R&D.

SystemVue/VSA M9390/81180A Wideband PSG DUT Infiniium scope 90K

Figure 5.2. SystemVue, in combination with best-in-class AWG and oscilloscope test families, provides a closed-loop, stimulus/
response modeling, and verification platform up to 32 GHz wide. It enables a versatile and cost-effective UWB system-level
approach in R&D.

Applying Ultra-Wideband Technology in Radar and EW Systems 59

UWB Waveform Creation
Carefully designed and optimized waveforms can be created using SystemVue running within the Infiniium
90000 X-Series oscilloscope, or on an external PC. SystemVue plays a critical role in UWB waveform
creation by providing a workspace with open, parameterized signal processing diagrams for LFM, pulse,
and noise UWB radar signal generation. Transmitter signals can be generated for Linear and Nonlinear
FM pulses and coded signals; sort pulse signals with Gaussian windowing; and noise UWB radar.
Radar target return signals with radar cross-section (RCS), clutter, jamming, and interferers can also
be generated.

Before downloading to test equipment, the simulated waveforms can be verified in SystemVue for
conformance to both frequency- and time-domain specifications. SystemVue also enables engineers to
incorporate custom signal processing intellectual property (IP) and create custom signals that integrate
C++ dynamic link libraries, MATLAB models, VHDL, and test vector data files.

Example: LFM UWB Transmitter/ Receiver Signals

To better understand how easily UWB radar transmitter and receiver signals can be generated and
measured, consider the example of LFM UWB transmitter/receiver signals. Using the test platform shown
in Figure 5.2, an LFM UWB transmitter signal is created with 1 GHz bandwidth, 1 microsecond of pulse-
width, and a 10-microsecond repetition interval. It is generated using SystemVue with the 81180 AWG,
the PSG, and the 90000 X-Series oscilloscope (Figure 5.3).

Figure 5.3. An LFM design in SystemVue (upper image) is used to generate a UWB signal. The final block in the
system-level schematic captures the simulation result and downloads the waveform data into the 81180A AWG using
the parameters shown (lower right). Then, the 81180 (not shown) repeats the generated signal for hardware testing

Applying Ultra-Wideband Technology in Radar and EW Systems 60

The LFM UWB transmitter signal is then measured using VSA software with a pre-stored configuration file.
The results are shown in Figure 5.4.

For receiver component test, the same equipment is used to create an LFM UWB receiver signal for a
target with a 100 meter range and a 20 m/s velocity. The 89600B VSA software is used to capture the
LFM UWB receiver signal. The VSA measurement is configured using a stored “setup” file which is quickly
recalled at the simulation runtime.

Note that for either or OFDM type transmitter signals, the impulse or OFDM UWB source objects must
first be activated in the SystemVue simulation to generate the respective UWB signals. Like the LFM
UWB signals, these alternate types of UWB signals can be measured using the VSA software.

Figure 5.4. Shown here is a 1-GHz LFM UWB signal. The upper left image shows the 1-GHz wide radar spectrum centered at
10 GHz, while the log magnitude envelope versus time is shown just below it. The signal phase is shown on the upper right
graph. The 1 GHz-wide LFM chirp is displayed on the lower right graph.

Applying Ultra-Wideband Technology in Radar and EW Systems 61

Summary of Results
Designing, verifying, and testing UWB radar systems requires precisely controllable UWB signals.
Unfortunately, generating and measuring the required UWB radar signals is no easy task. Keysight’s
SystemVue with a wideband AWG and the PSG signal generator provides a “simulatable” UWB radar
reference design and working waveforms that engineers can use to test and troubleshoot UWB
radar transmitter and receivers, thus reducing development effort. When Keysight’s 90000 X-Series
oscilloscope is added, radar transmitter and receiver measurements can also be performed, completing
a full “roundtrip” signal processing path for radar architecture validation up to 32 GHz wide. Together
these instruments create an interactive UWB test system that can be used to optimize UWB system
architectures conveniently and cost-effectively, as well as to verify individual RF and baseband
components over challenging bandwidths and signal conditions.

REALISTIC ULTRA-WIDEBAND RADAR SIGNAL GENERATION USING

PATHWAVE SYSTEM DESIGN (SYSTEMVUE)
https://www.keysight.com/see/radar-ew-system-design

Applying Ultra-Wideband Technology in Radar and EW Systems 62

Chapter 6
Dealing with the Complexity of Phased-Array Systems
Phased array is widely used in modern radar systems for rapid multi-target search and track
operations, as well as to achieve higher resolution, and better detection performance. Despite
these enviable benefits, when developing phased array radar, many issues may be encountered.
For modern engineers, that often means a myriad of test challenges, such as finding a way to
improve performance while also reducing the high cost of Transmit/Receive (T/R) modules with
Direct Digital Synthesizers (DDSs), digital-to-analog converters (DACs), and Analog-to-Digital
Converters (ADCs). Also of concern to the engineer is finding a way to work effectively with the
entire development team—the system architect, the RF team, and the signal processing team.
Additionally, calibration of the T/R module can be difficult, not to mention time-consuming and
expensive. Addressing these challenges demands an appropriate method of designing and testing
phased-array radar systems; one that streamlines the R&D lifecycle so that faster, cheaper, and
better phased-array radar systems can be achieved.
Phased-Array Radar Design: The Basics
There are two types of phased-array radar systems: passive and active (Figure 6.1). In a passive system,
a baseband source is connected to a single large Transmitter (Tx) with a High-Power Amplifier (HPA). The
Tx is connected to a beamformer followed by the antenna unit, the return signals of which are connected
to a single receiver (Rx) and subsequently, to the baseband receiver. In passive systems, the signal loss
between radiating elements and the T/R can be quite large. However, because passive antenna systems
have a central Radio Frequency (RF) source, developing a radar system based on a Passive Electronically
Scanned Array (PESA) is a straightforward process. The same cannot be said of radars based on an
Active Electronically Scanned Array (AESA). In contrast to a PESA radar, AESA devices have T/R modules
containing small Tx and Rx designs located behind each radiating element, and the baseband source is
connected to the beamformer. Transmitter power is distributed through many small PAs to the antennas,
while the baseband receiver receives signals through antennas in many small Low Noise Amplifiers
(LNAs). In an active system, the signal loss between the PA/LNA and the radiating element is much
smaller than in a passive system. Electronic scanning is therefore used, which enables faster, more flexible
searching. However, because each module contains its own RF source, development of AESA radars is
substantially more complex.

Figure 6.1. Two types of phased-array radar, passive-array and active-array antenna systems, are shown here.

Dealing with the Complexity of Phased-Array Systems 64

The Platform Solution
Dealing with the complexity of AESA radar development, while also addressing the traditional problems
and challenges associated with developing a phased-array radar requires a platform solution that enables
effective design and tests at every stage of the development process (Figure 6.2). The ideal platform
solution relies on simulation as its foundation and features several key characteristics, including cross-
domain simulation with RF and Electromagnetic (EM), as well as the ability to measure both 3D and 2D
antenna patterns. The measured antenna patterns, coupled with Tx measurements (e.g., waveform,
spectrum, and time-side-lobes) and Rx measurements (e.g., detection rate and false alarm rate) can be
used for performance validation.

The platform solution also offers trade-off analysis, T/R module and antenna unit failure analysis, and
adaptive algorithm creation support. It features links to test equipment (e.g., a signal generator, arbitrary
waveform generator (AWG), and signal analyzer) for hardware testing, along with support for integrated
testing. The links allow data to be downloaded to an AWG for testing RF signals and hardware signals
to be acquired and sent back to simulation for post-analysis. A prime example of this platform solution is
PathWave System Design, (SystemVue).

Benefits from AESA

prototyping platform
• Overcome test challenges for
calibrations in active antennas
• Reduce product Cost by
Trade-Off analysis

Figure 6.2. This platform solution for AESA radar relies on simulation at its core, in this case, the SystemVue
simulation software from Keysight Technologies, Inc.

Dealing with the Complexity of Phased-Array Systems 65

Key Models in the Platform
• Direct Digital Synthesis (DDS)
DDS is a key model for any digital radar and is frequently used in AESA radar for T/R module design
(Figure 6.3). It is a digital radar source for generating digital waveforms, such as Continuous Wave
(CW), pulse, Linear Frequency Modulation (LFM) pulse, stepped pulse, and stepped LFM as seen
from the downloaded I, Q waveforms, LFM, CW, pulse, and LFM pulse.
• Target model
When evaluating receiver performance, the radar environment has to be considered, which makes
creating a practical target model very important (Figure 6.4). While other commercial radar simulation
products rely on an ideal radar equation for this model, SystemVue offers a much more practical
target model.

Figure 6.3. A DDS model is shown here.

Moving target model

–  Earth Effect
–  Atmospheric loss
–  More RCS types
–  System_Loss
–  Ground refleciton
–  Polarization
–  Dielectric effection
–  Trajectory

Figure 6.4. Illustrated here is a target model.

Dealing with the Complexity of Phased-Array Systems 66

The radar environment includes terrain and sea surfaces, the atmosphere (including precipitation), and
the ionosphere. These conditions may degrade radar observations and performance by producing clutter
and other spurious returns, signal attenuation, and bending of the radar-signal path, including radar cross
section (RCS), Doppler, delay, attenuation, and propagation effects.

Even though “free space analysis” may be adequate to provide a general understanding of a radar
system, it is only an approximation. To accurately predict radar performance, the free space analysis
must be modified to include the effects of the earth and its atmosphere. Note that radar clutter is not
considered as part of this analysis because it almost always is assumed to be a distributed target that
can be dealt with separately by the radar signal processor.

• Clutter model
A clutter model is used to model the unwanted echoes in a radar system (Figure 6.5). The echoes are
typically returned from ground, sea, rain, animals/insects, chaff, and atmospheric turbulence, and can
cause serious performance issues with radar systems. Clutter can be best modeled using a statistical
approach that combines the probability density function (PDF) for clutter amplitude and clutter power
spectrum density (PSD). The PDF is used for the time-domain statistical property description, while
the PSD is used for the frequency-domain description. Both are suitable for describing the effects of
the radar environment. The K-clutter model is another important statistical model and is used for sea
and round clutters.  

Figure 6.5. Shown here is a clutter model.

Dealing with the Complexity of Phased-Array Systems 67

 • Array antenna model
The antenna pattern can be specified by the user using UserDefinedPattern or calculated based
on the size of the antenna and illuminating window function including Uniform, Cosine, Parabolic,
Triangle, Circular, Cosine Square, and Taylor. Array antenna models for the Tx and Rx allow the
user to specify the arbitrary geometry of the antenna pattern using the AntennaPatternArray in the
UserDefinedPattern (Figure 6.6). The ThetaAngleStart and ThetaAngleEnd give the scope of the
elevation angle, while the PhiAngleStart and PhiAngleEnd give the scope of the azimuth angle.
AngleStep is the value of the angle step for the user defined pattern.
• Beamforming model
Consider a uniform line array. Through signal processing, spatial filtering for interference can be
archived. Propagation can form a response pattern with higher sensitivity in desired directions. One
of the key technical problems of phased arrays is beamforming. To sum all signals from the array
antenna coherently, the time delay of the signal received by the antenna element at the position has
to be compensated.
When Τ = θ, the channels are all-time aligned for a signal from direction θ. Wi is beamformer weights.
Using Wi with each element allows the signal to point in any direction. The gain in direction θ is Σwm.
It is less in other directions due to incoherent addition.
A beamforming model can be used to help ensure the beamforming technique is optimally
implemented in a phased-array radar system (Figure 6.7).

Figure 6.6. An array antenna model is shown in this figure.

Dealing with the Complexity of Phased-Array Systems 68

Figure 6.7. A beamformer model is shown here.

Antenna Pattern Measurements

Both 3D and 2D antenna pattern measurements can be implemented for array systems.

In the design in Figure 6.7, a signal source is followed by a Tx beamformer to specify a 16-x-16
rectangular array with beam direction at Phi=0 and Theta=0. Through a T/R module, a Tx array antenna
model is used to send out the defined signals. Then, an ideal transmitter is used, followed by a sink to
collect the transmission data. Next, post processing is used in an equation block and MATLAB 3D plots
are used in 3D and 2D (Figure 6.8).

–  Source
–  Tx Beamformer
–  T.R Module
–  2D 3D Antenna
–  Pattern Measurements
–  MATLAB® Co-Simulation

Figure 6.8. An example of array antenna pattern measurements.

Dealing with the Complexity of Phased-Array Systems 69

RF Co-Simulation
A radar cross-domain simulation example for the Tx design and verification is shown in Figure 6.9.
The source and measurement are in the system-level data flow. The Tx is linked to the circuits with
X-parameters using SystemVue’s Co-Sim capability. Waveform Composer is used for a Tx source in which
a frequency hopping radar source with different frequencies at 1.3,1.5 and 1.7GHz is created. Using this
setup, the engineer can design and verify a complex RF transmitter with frequency hopping signals.

RF Co-Simulation Challenge
Transmitter Test
How to account for
accurate RF at the
system-level

Figure 6.9. An example of a RF co-simulation transmitter test.

Dealing with the Complexity of Phased-Array Systems 70

System Simulation Example Using SystemVue
As an example, a phased-array radar designed in SystemVue is shown in Figure 6.10. The design
includes an LFM source, Tx beamformer with T/R modules, array antenna with 16 sub-arrays,
radar moving target RCS, radar environments (e.g., clutter and jamming/Interference), Rx receiver,
Rx beamformer with T/R modules, and receiver signal processor for pulse compression, MTI, and
MTD measurements.

Figure 6.10. A phased-array radar simulation example.

After running the simulation, the following waveforms can be observed, as shown in Figure 6.11: figure A
is an LFM transmission signal, figure B is the received signal (return target plus clutter) hidden in strong
clutter and noise, and figure C is the return target with pulse compression hidden; MTI and MTD are
recovered and detected. In figure D, the return target through the signal processor (MTI and MTD)
is recovered.

In Figure 6.12, two 3D displays are plotted to show the detected signal in the Range-Doppler plane. The
first one shows the detected signal without any signal processing, while the second shows the signal with
signal processing.

Using the template in Figure 6.10, the user can quickly put together a phased-array radar system
for algorithm design, with a phased-array antenna and advanced signal processing. A cross-domain
architecture is also supported, and complex environments and advanced measurements are considered.

This allows the user to insert a custom algorithm and re-use their Intellectual Property (IP) for design
and validation.

The template is easy to use because the design schematic, measurement results, and estimated
parameters are all shown at the top level of the screen, eliminating the need for the engineer to have to
track down results from various places. Key specified parameters can be easily modified using sliders at
the same top level. The engineer simply makes one click and the results populate the screen, including a
3D plot, measurements, and parameter estimations.

To create a custom design, engineers need not start from scratch. Instead, the example design in Figure
6.10 can be used as a template. The user simply modifies the template by changing parameters and
replacing models with existing models in the radar library or by importing custom models into the design.

Dealing with the Complexity of Phased-Array Systems 71

Figure 6.11. The phased-array radar simulation results.

Figure 6.12. The phased-array radar simulation results in a 3D plot.

Regardless of the radar system-level test solution utilized, core software is needed to integrate all test
software and hardware and to automate the test.

SystemVue integrates all test instruments together as a test system that provides complex radar test
signals with environment scenarios to the Device-Under-Test (DUT), to capture DUT outputs and then
synchronize signals, post-processing the result to extract more information and obtain more advanced
measurements, such as detection rate, false alarm rate, and imaging analysis. Without integration
and synchronization, each instrument would function on its own, making it impossible to perform
complex tests.

Dealing with the Complexity of Phased-Array Systems 72

Besides the single-channel test, multi-channel test for phased array and MIMO radar is possible using
SystemVue’s signal downloader to a Vector Signal Generator (VSG) or multichannel VSG like the M9381A
for a MIMO source. A VSA link to signal analyzers/scopes like the M9703 can be used for the MIMO
receiver test. The basic test system structure is shown in Figure 6.13.

Figure 6.13. Shown here is an integrated test system using SystemVue.

Dealing with the Complexity of Phased-Array Systems 73

Developing a phased-array radar system is challenging, especially when it involves an AESA device. In
this case, the engineer requires a much more integrated solution with a wide breadth of functionality. Use
of a platform solution that relies on simulation as its foundation now offers engineers an effective strategy
for attacking the challenges they face when designing, verifying, and testing phased-array systems. This
approach not only reduces design cycles but also significantly reduces cost.

SystemVue is a prime example of a platform solution. There are several key benefits to using it to design
and test phased-array systems. For example, trade-off analysis can be used to significantly reduce cost.
Adaptive algorithms can be used to fix amplitude/phase errors for calibration purposes. SystemVue
also provides emulation environments that account for clutter and Interference. Lastly, validation can be
performed based on measured antenna patterns, Tx measurements such as waveform, spectrum, time-
side-lobes; and Rx measurements such as detection rate and false alarm rate, which helps reduce the
design cycle.

Using the design templates provided in the model based SystemVue platform users can:

• Quickly put together a new system-level proposal by creating, with a higher level of confidence, an
integrated RF and DSP architecture design.
• Easily integrate IP written in different languages (e.g., C++, MATLAB, ADS, and HDL) at the system
level for a radar/Electronic Warfare (EW) system.
• Easily create complex radar/EW scenarios and verify systems with environment scenarios (e.g.,
clutter, jamming/deception, interference, and RCS) to meet complex system specifications and
perform virtual flight testing to reduce the high cost of the field test.
• Utilize an integrated test system.

HOW TO DESIGN PHASED ARRAY SYSTEMS

https://www.keysight.com/see/radar-ew-system-design

Dealing with the Complexity of Phased-Array Systems 74

Chapter 7
Generating Signals for Radar and EW Design and Verification
Modern radar systems use more complex signal formats working in wide or ultra-wide bandwidths,
and operating in different frequencies (e.g., X, Ku, and Ka bands). As we have seen in the previous
chapters, they also use advanced digital signal processing techniques to disguise their operation
and overcome strong clutter and jamming in their environment. Addressing this complexity requires
a generation of realistic test signals and system-level scenarios that can be used to create and
verify the radar signal processing algorithms. While dedicated hardware simulators and field testing
are typically used to generate these test signals, both are costly, time-consuming, and apply later in
the design process. This chapter presents a less expensive option for generating test signals early
in system development. This approach to system-level design and verification uses the Keysight’s
PathWave System Design (SystemVue) environment, a Radar Model Library, and commercial
off-the-shelf test equipment for the generation of continual and pulse radar waveforms, for both
algorithm and hardware verification.

Generating Signals for Radar and EW Design and Verification 75

Design Problem
Radar signal processing algorithms play a critical role in advanced radar, especially high-performance
multi-mode systems. Radar designers require custom reference waveforms with precise amounts of
impairment and field conditions when testing or verifying components (both baseband and RF). The signal
scenarios are used both for troubleshooting and block-level diagnostics, as well as for initial platform
integration and validation.

Unfortunately, signal processing algorithm creation is complicated, with all resulting algorithms needing
to be verified in the complex external environment. Greater interaction between algorithms and signal
sources provides a realistic radar environment that allows early maturity and confidence in the radar
system. Creating the advanced algorithms, therefore, requires the availability of a sufficient set of models
for the various radar elements and functions, including signal generation, transmission, antenna, T/R
switching, clutter, noise, jamming, receiving, signal processing, and measurements.

Common Approaches to Radar System Design

The two methods commonly employed by radar designers to generate the required signals for algorithm
creation are accelerated, dedicated hardware simulation, and direct field testing.

Hardware simulation
Scenario-based verification is typically performed using expensive, high-end hardware simulators that
create real-time scenarios of realistic fidelity. Unfortunately, such simulators are not available early enough
in the design process (during the algorithm stage before hardware implementation), to provide meaningful
insight for the algorithm developer. They also don’t integrate well with model-based design practices or
provide the level of scripting needed in early development for low-level, block-level verification. Because
simulation takes place later in the design cycle, making design changes at this stage can be both costly
and time-consuming. Design engineers can benefit from earlier algorithmic validation of their key ideas.

Field testing
Another way to generate the required signals is to physically go into the field to make environmental
measurements of the radar hardware operating under realistic conditions. While there are many reasons
for utilizing this approach, because it comes late in the design process it rarely connects to the original
algorithmic development environment. In addition, this method is costly, sometimes inconvenient,
and cumbersome.

A More Practical, Less Expensive Option

For early development, a robust and inexpensive alternative uses a convenient software simulation-
based radar library to generate the reference waveforms needed for radar signal processing algorithm
development. Early access to these signals enables the design of superior system architectures
and facilitates early hardware verification at different layers of abstraction, as the model-based
design matures.
Generating Signals for Radar and EW Design and Verification 76
The Keysight radar library application
As can be seen from Table 7.1, the Keysight W4521 Aerospace / Defense Library is an advanced
simulation block set of over 35 highly parameterized primitive blocks and higher-level reference
designs. It can be used for modeling different types of radar systems, creating radar signal processing
algorithms, evaluating the system’s performance, and providing proof-of-concept designs. The block
set and its example workspaces serve as algorithmic and architectural reference designs to verify radar
performance under different conditions. These include target and radar cross section (RCS) scenarios,
clutter conditions, jamming (intentional), and environmental interference, and the effect of various receiver
algorithms. It is ideal for radar designers who need to generate precise signals for algorithm and hardware
verification or study the performance of their radar systems under various conditions.

Signal sources LFM – Linear FM wave generator

NLFM – Nonlinear FM wave generator
BarkerCode – Poly-phase code wave generator
FrankCode – Frank code wave generator
ZCCode – Zadoff–Chu code wave generator
MatchedSrc – Generates the matched source signal for pulse
compression
Signal processing Detector – Video signal detector
FFT – Complex fast Fourier transform
PC – Pulse compression processing
PD – Pulse Doppler processing
MTI – Moving target indication
MTD – Moving target detection
CFAR – Constant false-alarm rate process
Window – Windowing for sidelobe control
Transmitter CICInterp – Interpolation with cascaded CIC filters
DUC – Digital up-converter, baseband to intermediate frequency
UpSample – Up sampler with poly-phase filter
Tx – Transmitter front end
Receiver CICDecimate – decimation with cascaded CIC filters
DDC – Digital down-converter, intermediate frequency to baseband
DownSample – Down sampler with poly-phase filter
Rx – Receiver front end
Environment RCS – Radar cross-section modeling
Target – Target modeling, including RCS, Doppler effect, delay,
attenuation
Clutter – Clutter modeling
Antenna T x Ant – Transmitter antenna
Rx Ant – Receiver antenna
Measurement Pd Measurement – Detection probability estimation
Pf Measurement – False-alarm rate estimation

Table 7.1. The W4521 radar model library block list

Generating Signals for Radar and EW Design and Verification 77

The W4521 works within the SystemVue system-level design environment. This is an open modeling
environment focused on physical-layer baseband/RF architectures that replace general-purpose digital,
analog, and math environments, and co-simulates with HDL for embedded hardware design flows.
Together SystemVue and the W4521 create a system-level platform for design and verification that meets
the requirements for signal processing algorithm creation (Figure 7.2).

Generate waveforms - Return signals with clutters + jamming

Signal
g ggenerators

Signal
processor
hardware

Signal analyzers
logic analyzers, scopes

Figure 7.2. The W4521 Aerospace/Defense Library and SystemVue system platform.

Generating Signals for Radar and EW Design and Verification 78

The platform provides a user-friendly algorithm modeling and debug environment that supports a variety
of languages (e.g., C++ and math language), as well as an algorithm design and verification environment
for signal generation and performance measurements (Figures 7.3 and 7.4). From the block diagram in
Figure 7.3, the main models include signal source, transmitter, antenna, T/R switch, RCS, clutter,
jammer, receiver, signal processors, and measurements. SystemVue enables the creation of new and
customized algorithms for the specific applications of the radar system under design. Many of the
available sub-libraries are listed.

Tx waveform T/R Antenna Radar Radar signal

library component library environment processing
– Radar pulse – DAC – Antenna models – Target – Digital pulse compression
generator – PA – T/R antenna array – Clutter – Moving Target Indication (MTI)
– Coherent signal – Up converter – Antenna propagation – Jammer – Moving Target Detection (MTD)
generator for single – Filter – Interference – Constant False Alarm Rate (CFAR)
or multiple channel – DDS – Digital beamformer
– LNA – Space-Time Adaptive Processing (STAP)
– Down converter
– ADC

Figure 7.3. SystemVue combined with the optional W4521 Aerospace/Defense Library is a platform for radar
system design.

Generating Signals for Radar and EW Design and Verification 79

 PSA/MXA/VSA/PXA
ESG/PSG/MXG/PXB Signal processor LA, scope
hardware

Figure 7.4. A platform for radar verification.

As shown in Figure 7.4, radar components can be connected to instruments and test hardware based on
the created algorithm.

To better understand how the W4521 Aerospace/Defense Library and SystemVue platform generate radar
test signals, consider the test setup in Figure 7.5. Here, an interface model (Sink) in SystemVue connects
to one of the Keysight ESG/PSG/MXG families of vector signal generators. SystemVue generates radar
waveforms in simulation that are automatically downloaded at run-time into the instruments for use as RF/
IF test signals during hardware testing. Waveforms captured by the vector signal analyzer can then be
automatically returned to SystemVue. Acquired waveforms can be further processed in SystemVue.

Generating Signals for Radar and EW Design and Verification 80

Effectively, a smart simulation platform (SystemVue with the Aerospace/Defense Library) surrounded
by wideband stimulus/response equipment can be used to manually imitate missing hardware blocks,
to complete a working radar system. This allows system-level validation to continue at an earlier date,
based on partially implemented hardware. As real hardware becomes available, the simulation platform
withdraws to function more effectively with the test equipment to provide targeted radar signals for
testing. By working early within the radar block diagram, SystemVue serves a variety of R&D needs such
as design transitions from concept to working prototype, eventually assisting with hardware testing. This
serves an expanded function relative to the Keysight Signal Studio for pulse building software, which is
focused specifically on test signal creation using Keysight signal generators, for final hardware validation.

SystemVue DUT
Signal generator Signal analyzer

Figure 7.5. Platform for radar test signal generation.

The SystemVue platform provides an interface to a range of test equipment to help verify the implemented
hardware (compared to the original pure algorithm). Supported algorithmic reference sources
include radar signal generation with RCS, clutter, jamming, and Doppler frequency offset. Supported
measurements include waveforms, spectra, detection rate, and false alarm rate (FAR).

Some of the test equipment that can be connected to SystemVue includes signal analyzers, logic
analyzers, and scopes (e.g., Keysight’s Infiniium 90000 X-Series high-performance oscilloscope).

Generating Signals for Radar and EW Design and Verification 81

Radar Component Test Examples
To test radar components in a transmitter and receiver, the W4521 can be used to generate various types
of test signals including radar transmission signals such as LFM, NFM, Coded FM, return (RCS) signals,
and return signals with clutter. An example of return signals with clutter is shown in Figures 7.6 to 7.9.

Generate waveforms - Radar transmission (LFM) signals

Signal generators

Signal
processor
hardware

Signal analyzers
logic analyzers, scopes

Figure 7.6. This diagram shows how a LFM radar transmission signal can be generated in SystemVue and then
downloaded to Keysight’s ESG/MSG/PSG vector signal generator. The LFM transmission signal can be used for
radar component testing.

Generating Signals for Radar and EW Design and Verification 82

 Generate waveforms - Return (RCS) signals

Signal generators

Signal
processor
hardware

Signal analyzers
logic analyzers, scopes

Figure 7.7. This diagram depicts how to generate the radar receiver test signal. Users can use the SystemVue radar
Target model to specify target range, velocity, and radar cross-section models. The receiver signal can be used to
test radar receiver detection algorithms.

Generating Signals for Radar and EW Design and Verification 83

 Generate waveforms - Return signals with clutters

Signal generators

Signal
processor
hardware

Signal analyzers
logic analyzers, scopes

Figure 7.8. The resulting radar received signal with RCS and clutter describes the radar environment details.

Generating Signals for Radar and EW Design and Verification 84

 Generate waveforms - Return signals with clutters + jamming

Signal generators

Signal
processor
hardware

Signal analyzers
logic analyzers, scopes

Figure 7.9. Here, the radar received signal has not only RCS and clutter but jamming as well for describing the
radar’s complex environment details.

The W4521 can also be used to generate wideband and ultra-wideband signals using the setup shown
in Figure 7.10. In this case, SystemVue generates radar baseband I/Q data that is downloaded to the
N6030 or N81180 arbitrary waveform generator—a 4.2 GSa/s arbitrary waveform generator with a
12-bit vertical resolution for complex real-world signals. The wideband radar signals formed in the
N6030 or N81180 are then sent to the wideband PSG I/Q modulator inputs, which in turn sends RF
wideband signals to the DUT to test the RF radar components. Next, the output DUT is captured using
the Infiniium 90000 X-Series scope and sent back to the PC for either direct analysis using the VSA
software or for further analysis using SystemVue. A generated radar received LFM signal with RCS is
shown in Figure 7.11.

Generating Signals for Radar and EW Design and Verification 85

 PXI hyper board or
N6030/ DUT Infiniium scope
SystemVue/VSA (N81180) Wideband PSG

Figure 7.10. Test setup for wideband signal generation.

Figure 7.11. This wideband signal (radar received LFM signal with RCS) was
generated using SystemVue, the N6030, and PSG. The signal was measured using
VSA software.

Generating Signals for Radar and EW Design and Verification 86

The complexity of advanced radar systems puts added focus on the radar signal processing algorithms
which are critical to their development. Creating these algorithms requires a generation of test signals.
While this is typically done using high-performance hardware simulators or range testing, Keysight’s
W4521 Aerospace/Defense Library offers a more practical, timely, and drastically less expensive
alternative. When used with the SystemVue environment, this application provides the basic information
needed to create the required test signals earlier in the design cycle, enabling superior system
architecture design and facilitating early hardware verification at several layers of abstraction as a design
matures. Examples of generated test signals (return signals with clutter and radar received FM signal with
RCS) demonstrate the validity of this solution.

Generating Signals for Radar and EW Design and Verification 87

Chapter 8
Virtual Flight Testing of Radar System Performance
Taking a system-level approach to a complex design often requires upfront integration and
analysis, however, it pays in the long run by focusing time and engineering effort on “winning”
design strategies. In this chapter, PathWave System Design (SystemVue) is integrated with the
STK software from Analytical Graphics Inc. (AGI, an Ansys company) to take advantage of their
respective domain strengths in order to address difficult radar modeling and verification issues.

In addition, the SystemVue Aerospace/Defense library provides radar signal processing/domain IP

to render the final details of this particular system and a friendly interface for modeling and
test equipment.

Virtual Flight Testing of Radar System Performance 88

Problem Statement: Reducing Dependency on Flight Testing
Flight testing is the ultimate way to evaluate the performance of a radar and EW system. During the
actual aircraft flight, data such as Probability of Detection, Signal Strength, and clutter might be gathered.
While effective, this approach does pose several challenges. The operational cost of flight testing a radar
system using real aircraft can be over $100,000 per hour (Figure 8.1). Additionally, the results from one
flight to the next are not repeatable. Each flight is slightly different and getting enough flights in to be
statistically significant is simply too costly.

Aircraft testing and evaluation costs

Flight test Ground test
1200

1000

800
$M

600

400

200

0
t

les

el

er
af

or

or

or

nn

th
cr

pp

pp

pp
tic

O
tu
air

ar
su

su

su

d
st

st

in
g

st

st
te

te
in

W
te

te
er
ht

d
ht

d
ne

un
ig

un
ig
Fl

gi

ro
Fl

ro
En

Opportunities for lean thinking in aircraft flight testing and evaluation

Carmen F Carreras-Massachusetts Institute of Technology, June 2012
Figure 8.1. The high cost and time associated with flight testing is an incentive to explore alternative solutions.

While final operational verification may still be necessary for contractual or legal reasons, “virtual flight
testing” is a faster, more cost-effective alternative for earlier stages of R&D, such as algorithm and
countermeasures development. In a simulation, complex radar systems can be evaluated hundreds of
times in an hour, using the same or varied scenarios for each run (flight), and at significantly less cost than
a single hour on a flight range. By evaluating realistic flight-testing scenarios before or in place of physical
flight testing, engineers can validate electronic warfare algorithms earlier, saving both time and money.

Virtual Flight Testing of Radar System Performance 89

A Virtual Solution
A virtual flight test solution can be created by marrying the capabilities of Keysight’s PathWave System
Design (SystemVue). SystemVue is electronic-system-level design software that integrates modeling,
simulation, reference IP, hardware generation, and measurement links into a single, versatile platform
(Figure 8.2). It enables system architects and algorithm developers to innovate the physical layer (PHY)
of wireless and aerospace/defense radar and communications systems and provides unique value to
RF, DSP, and FPGA/ASIC implementers. The W4521 Aerospace/Defense library provides baseband
signal processing reference models for a variety of radar architectures. STK is a physics-based software
geometry engine that accurately displays and analyzes land, sea, air, and space assets in real or
simulated time. It can include the aircraft flight dynamics, terrain effects, and the aircraft’s 3D radar cross
section (RCS).

Simulation
Open modeling – Integration capability
– Existing modeling templates – Connection to leading RF EDA flows
– Custom models: C++,.m, HDL – Performance evaluation for RF & BB
– Model import: MATLAB, ADS, – Advanced measurement
SignalStudio, VSA, STK – Detection rate, flase alarm rate
– Recorded data System Vue – Dynamic range
– Advanced dataflow engine – Parameter estimation
– Co-simulation
– RF envelope simulation
– Model libraries
– Integration of SW, HW
– HDL simulation
– FPGA implementation

HW implementation HW test
– Existing modeling templates – Link to VSG/VSA/Scope/LA
– DSP algorithm creation – Integration/Controlling/Automation
– Fixed point simulation – Custom waveform generation
– HDL code generation – Advanced RF & BB measurements
– FPGA synthesis – Parameter estimation
– Troubleshooting

Figure 8.2. With SystemVue, PHY development is enabled across RF and baseband domains and that development
can easily transition from algorithms into hardware verification.

The basic STK process is to define a system link scenario with moving transmitter (Tx), receiver (Rx),
and interferer objects. The scenario is then analyzed to obtain system metrics as a function of time (e.g.,
range, propagation loss, RCS, noise bandwidth, and Rx signal strength). Almost everything in STK can
be controlled by third-party tools. However, the software has no inherent ability to process signals from
radar/communications applications through the dynamic environment link. Linking STK with SystemVue
allows arbitrary Tx/Rx radar/communications systems to be modeled with the STK dynamic environment
link characteristics. During virtual flight testing, SystemVue models the radar system including waveform

Virtual Flight Testing of Radar System Performance 90

generation, Tx and Rx non-ideal behavior, DSP and RF processing, and radar post-processing, while STK
models the flight scenario and signal path characteristics (e.g., path loss, Doppler, aircraft aspect RCS,
and atmospheric losses).

Virtual Flight Testing Example

To gain a clearer understanding of the interface between SystemVue and STK, and its application to
virtual flight testing, consider the 3D STK simulation scenario of a fighter sortie (Figure 8.3). In this
example, assume the sortie starts at 10,000 feet and is detected by radar. To try to get below the
radar, it dives down to do low-level terrain-following, sometimes successfully, sometimes not. The same
run can be repeated hundreds of times, with different radar or electronic countermeasure assets in
place as modeled by SystemVue, along with the terrain, aircraft (including 3D RCS), and the radar site
characteristics as modeled in STK.

STK: Virtual flight path, with terrain information

Doppler shift, delay, other Detection probability

dynamicfading parameters with STK link

LFM STK
Transmitter Receiver Signal processing Pd
link

SystemVue: TX/RX radar signal processing chain, with parameterized fading PROBABILITY
OF DETECTION
Figure 8.3. One application of the interface between SystemVue and STK is the ability to do virtual flight testing
of radar systems, including DSP, RF impairments, jamming, and interference as an aircraft encounters targets and
clutter along a virtual flight plan.

Virtual Flight Testing of Radar System Performance 91

As shown in Figure 8.4, a custom user interface can be easily implemented within SystemVue to make
repetitive tasks and complex measurements much easier to manage. Here, SystemVue creates a radar
waveform and passes it through a transmit chain to multiple target models (including jamming and added
clutter). The resultant RF waveform can then be input into an arbitrary waveform generator and introduced
into a receiver for performance validation. SystemVue also has tight integration with MATLAB, C++, and
HDL simulators so existing radar algorithms can also be integrated into the scenario. Measurement-based
data, such as a jammer profile or measured interference, could also be added into the simulation directly
through Keysight test equipment links.

Virtual Flight Testing Example

SystemVue (Keysight) STK (AGI)
Radar signal path models Radar signal path models

Measurements
Tx waveforms Targets

MATH
.m Antenna

User’s math, C++

Wideband AWG Link
HDL algorithms Clutter

Figure 8.4. In this multiple target signal emulation example, test entry comes from a custom user interface with
hardware text flavor. The user does not have to open a simulation schematic. This approach integrates both signal
generation and signal analysis.

Virtual Flight Testing of Radar System Performance 92

Flexible API Enables Custom Applications and Measurements
Linking the SystemVue and STK solutions allows for quick and repeatable validation of multiple realistic
radar system scenarios. These scenarios can be evaluated in place of physical flight testing or, in cases
where operational flight testing is unavoidable, they can be evaluated beforehand to ensure they make the
most effective use of resources.

Some applications of virtual testing:

• Evaluate new jamming techniques or threats

• Inject multiple dynamic emitters and targets into your scenarios
• Allow various types of jamming based on a defined set of criteria for dynamic operation

• Modeling and evaluation of cross-domain effects, such as automatic gain control

• Include unintended interference from commercial wireless networks

The interface presented in this chapter started with the commercially available SystemVue and STK
environments and then used their application programming interfaces (APIs) to link them together.

When it comes to testing radar system performance, extensive flight testing using physical aircraft is a
prohibitively expensive and time-consuming proposition. Virtual flight testing, made possible by the flexible
interfaces between the SystemVue and STK software tools, now offers an economical alternative for R&D
validation. This allows measurement hardened algorithms to be deployed quickly, and a minimum of true
operational testing to be done with greater confidence, to save costs. By closing the loop between lab-
based virtual testing (simulation and test equipment) and operational testing, virtual testing can be made
even more effective.

BRINGING REAL-WORLD SCENARIO MODELING TO

AEROSPACE SYSTEMS DESIGN
https://www.keysight.com/see/radar-ew-system-design

Virtual Flight Testing of Radar System Performance 93

Section 3

RF Component Test
and Verification
Introduction
As radar and EW systems become ever more sophisticated, their demanding performance
requirements flow downstream to produce challenging specifications at the component level,
with a need to characterize at higher frequencies, wider bandwidths, lower noise levels, and with
complex signals.

This section addresses the measurement fundamentals of Distortion, Noise, Phase Noise,
and S-parameters. We also cover the “Active-Hot” characterization of nonlinear devices such
as amplifiers, introducing the concept of X-parameters and the benefits. Finally, we look at
characterizing materials that may plan to be used in radomes or for RF absorption, in and around
your radar and EW system.

VNA PORTFOLIO SPOTLIGHT - CHARACTERIZE BEYOND S-PARAMETERS

WITH PROVEN HARDWARE AND ADVANCED SOFTWARE
https://www.keysight.com/see/radar-ew-component-test
Chapter 9
Figures of Merit Aid Amplifier Distortion Assessment
Distortion not only degrades transmitter performance but also interferes with other receivers.
There are two major types of nonlinear distortion measurements — harmonic and intermodulation
distortion. Furthermore, no system offers unlimited gain and it is important to understand the
compression characteristics of amplifiers in particular, where the saturation point of the output is an
important figure of merit.

Harmonic Distortion
The amplitude transfer characteristics of a circuit or device cannot precisely track the input signal. The
amplitude shifts generate higher frequency components at integer multiples of the input signal.

Using a continuous wave (CW) tone as an input signal and measuring the output signal with a signal
analyzer is the most straightforward method for measuring harmonic distortion; see Figure 9.1. A DUT
might be an RF amplifier or mixer.

Fi < Fc < 2Fi

Fi Fo
Signal
LPF DUT
Analyzer
Signal
Generator
Fi Fi = Fo

Amplifier
Amplitude

∆2
Fi Fo
∆3
Mixer

Frequency

Fo 2Fo 3Fo
Figure 9.1. Harmonic distortion measurement setup

Figures of Merit Aid Amplifier Distortion Assessment 96

The signal analyzer sets to zero-span, which enables a time-domain power measurement to measure
the power level at the fundamental and harmonic frequencies as shown to the right of Figure 9.2. For a
higher-order harmonics measurement, you can choose a higher dynamic range signal analyzer.

Figure 9.2. Harmonics measurement with Keysight signal analyzer

Third-Order Intermodulation
Two-tone, third-order intermodulation (TOI) distortion is a common test for RF distortion measurements.
When two or more signals are present in a nonlinear system, they can interact and create additional
components at the sum and difference frequencies of the original frequencies and at sums and
differences of multiples of those frequencies. Figure 9.3 below shows the two-tone third-order
Intermodulation measurement setup. The device under test could be an amplifier or a mixer.

F1

Fo
SG #1 Signal
DUT Analyzer
F2

SG #2
PF1 PF2

Plower_intermod Pupper_intermod
Amplifier
Amplitude

F1+F2
2F1-F2 2F2-F1
Mixer ...
3F1-2F2 3F2-2F1
Frequency
F2-F1 F1F2 2F1 2F2

Figure 9.3. Two-tone intermodulation distortion measurement setup

Figures of Merit Aid Amplifier Distortion Assessment 97

F1 (lower tone) and F2 (upper tone) are frequencies of the two test tones from two signal generators. The
two tones injected into the system must be free from any third-order products. The third-order distortion
products occur at frequencies 2F1-F2 and 2F2-F1 (red notations), which are the closest distortion products
to the original two test tones. Removing the closest distortion products with filtering is difficult. In a
communication system, the distortion products can be interfering signals to the adjacent channels.

The definition of TOI level:

TOIlower (dBm) = PF2/2 + PF1 - Plower_intermod/2

TOIupper (dBm) = PF1/2 + PF2 - Pupper_intermod/2

For test efficiency, a vector signal generator alone can be used to generate two test tones using the
internal baseband generator to save costs. Keysight offers an advanced correction routine that can
suppress distortion products generated by the signal generator itself or an external pre-amplifier. Swept
IMD measurement of amplifiers and frequency converters can also be performed using a VNA with an
internal combiner and two internal signal generators. Wide receiver dynamic range and high-quality signal
sources minimize the measurement errors found in traditional setups.

Figure 9.4. PNA-X with S93087B IMD measurement simplifies setup and provides guided calibration.

A third RF source can be added to the 4-port PNA with the same low-phase noise performance as the
two internal RF sources. You can use it as the local source of the DUT for two-tone frequency converter
measurements or two-stage frequency converter measurements. It eliminates the need for an external
signal generator. You can also use it as an independent analog signal generator. For example, it can
be used as the reference signal for a DUT that requires the external reference clock signal, and as the
reference signal of another measurement instrument.

The 4-port PNA-X can be equipped with extremely low phase noise performance on all internal signal
generators making it ideal for IMD measurements where the two-tone stimulus signals must be close to
each other.

Figures of Merit Aid Amplifier Distortion Assessment 98

 3rd Source

IMD IMD
Input Output

Figure 9.5. Two-tone IMD mixer measurement utilizing three high-performance analog signal generators built into
the 4-port PNA-X

Compression
The most common measurement of amplifier compression is the 1-dB compression point. This is defined
as the input power (or sometimes the corresponding output power) that results in a 1-dB decrease in
amplifier gain referenced to the amplifier’s small-signal or linear gain.

The example in Figure 9.3 shows an amplifier’s output versus input power measured at a single frequency.
The amplifier has a linear region of operation at which gain is constant regardless of power level. The gain
in this region is called small-signal gain and is proportional to the slope of the power response.

As the input power continues to increase, the point on the curve at which amplifier gain begins to
decrease defines where the compression region begins. The amplifier’s output is no longer sinusoidal in
this region, and some of the output appears in harmonics rather than only in the fundamental frequency
of the signal. As input power is increased, even more, the amplifier becomes saturated, and output
power remains constant. At this point, the amplifier’s gain drops to zero, and increases in input power will
typically not produce increased output power.

Figures of Merit Aid Amplifier Distortion Assessment 99

To measure an amplifier’s saturated output power over a power sweep, the stimulus must provide
sufficient output power to drive the amplifier into saturation. A booster amplifier may be needed at the
input of high-power amplifiers to achieve saturated conditions.

Saturated
output power
Output Power (dBm)

Compression
region

Linear region
(slope = small-signal gain)

Input Power (dBm)

Figure 9.6. Power sweeps characterize the compression region

HIGH-PERFORMANCE SPECTRUM ANALYSIS FOR

PNA VECTOR NETWORK ANALYZERS
https://www.keysight.com/see/radar-ew-component-test

Figures of Merit Aid Amplifier Distortion Assessment 100

Chapter 10
Determine Your Optimal Noise Figure Measurement Method
Noise, and specifically signal-to-noise ratio (SNR), is a fundamental issue in wireless receivers.
High noise levels will limit system performance. A sensitive receiver means you can detect
lower-level signals, therefore a radar can maintain overall performance whilst transmitting at
lower power, reducing size, weight, and power requirements. Noise added at an early stage can
never be removed, therefore the gain and noise figure at the front-end of a receiver can dominate
overall sensitivity.

Determine Your Optimal Noise Figure Measurement Method 101

Definition of Noise Figure
The noise figure F of a network is defined as the ratio of the signal-to-noise power ratio at the input to the
signal-to-noise power ratio at the output.

S1 / Ni
F=
SO / NO

Thus, the noise figure represents the decrease (or degradation) in the signal-to-noise ratio as the signal
goes through a device or network. A perfect amplifier would amplify the noise at its input along with the
signal, maintaining the same signal-to-noise ratio at its input and output (the source of input noise is often
thermal noise associated with the earth’s surface temperature or with losses in the system). A realistic
amplifier, however, also adds some extra noise from its components and degrades the signal-to-noise
ratio. A low noise figure means that very little noise is added by the network.

Noise figure can be applied to both individual components such as a single transistor amplifier, or a
complete system such as a receiver. The overall noise figure of the system can be calculated if the
individual noise figures and gains of the system components are known.

Noise Figure Measurement Methods

Two techniques are generally used for measuring noise figures: the Y-factor method and the cold-source
method. The Y-factor method is the most common for RF and microwave frequencies (Figure 10.1). To
use this technique, connect a switchable, calibrated noise source to the DUT input and connect a noise
figure analyzer or signal analyzer to the output to measure the resulting noise, and then calculate the ratio.
A separate calibration measurement directly measures the output of the noise source in its on (“hot”) and
off states, and measures the gain of the DUT.

+28V

Noise
Source

DUT

Figure 10.1. In the Y-factor method, the central element of the noise source is a diode, driven to an avalanche
condition to produce a known quantity of noise power. The diode is not a dependable 50 Ω impedance, so it is
often followed by an attenuator to improve impedance match with the presumed 50 Ω DUT.

Determine Your Optimal Noise Figure Measurement Method 102

The cold-source method uses a vector network analyzer (VNA) with a two-port connection to the DUT.
A separate noise source is not required and a single connection to the DUT is sufficient for the entire
measurement. Both methods are capable of yielding accurate measurements for receiver design and face
challenges in some measurement conditions. Generally speaking, these challenges come in two forms:
accurately compensating for sources of error and separating noise in the measuring receiver from noise
originating in the DUT. For this chapter, we have limited our discussion to amplifiers as the DUT.

The most accurate measurements, especially those made under unfavorable conditions, are made using
the cold source method. This method is especially valuable for millimeter-frequency measurements or
when the input and output match of the DUT is poor—that is, substantially different from the 50 ohms
assumed for instruments, cables, and accessories.

The vector calibration techniques of a VNA can account for the multiple potential mismatch errors in the
measurement configuration. This reduces what is normally the largest single source of error in noise figure
measurements. Impedance mismatches result in erroneous power measurements, which directly affect
noise figure calculations.

Common sources of mismatch include the use of wafers, probes, or fixtures (or any type of non coaxial
connection); mismatch between the DUT and noise source or analyzer; and test system signal switching.
Mismatch also generally degrades as frequency increases, so the cold-source method is often the best
approach for millimeter-frequency measurements.

In practice, though, most noise figure measurements are made at RF and microwave frequencies, with
coaxial connections and a reasonable impedance match between the noise source, analyzer, and DUT.
For these measurements, cost and convenience lead many RF engineers to choose the Y-factor method.
It takes advantage of the signal analyzer that is more likely to be on their bench. These are usually less
expensive than VNAs and can provide a good combination of accuracy and measurement cost.

With these tradeoffs in mind, it important to pay some attention to the unfavorable conditions that can
make measurements difficult or substantially increase error:

• DUTs with a combination of low gain and very good noise figure: Such devices will generate very
little incremental noise. This may result in a signal at the analyzer input that is difficult to measure
accurately because it is very close to the analyzer’s noise floor.
• Conducted or radiated interference: Analyzers have no way of separating power that is due to
either conducted or radiated interference. If possible, measurements should be made in shielded
enclosures, with mobile phones and networks excluded, and on battery power, if practical.
• Complex or lengthy connections or adapters: Cabling and adapters can spoil impedance match and
attenuate the noise power that is supplied or measured. Custom cabling, which eliminates the need
for adapters or extra length, can be an inexpensive way to improve measurement performance and
reduce error.
• Inconsistent connections between calibration and measurement steps: Cable and connector care
and connection techniques, including proper torque, are especially important.

Determine Your Optimal Noise Figure Measurement Method 103

 • Devices that change temperature and drift during measurements: Noise figure measurements and
uncertainty calculations assume static parameters. Any drift (or other changes) after calibration or
during the measurement process will have a direct effect on accuracy.
• Finally, external or internal preamplifiers can improve noise figure accuracy by improving the noise
floor of the measuring instruments. To be beneficial, these preamplifiers should have extremely low
noise figures over the frequencies of interest.

Summary of Measurement Solutions

Keysight offers three types of solution platforms: a dedicated noise figure analyzer, signal/spectrum
analyzers, and vector network analyzers. The benefits of each are outlined below.

Noise figure analyzer (NFA)

The NFA Series is made exclusively for accurate noise figure measurements and uses the Y-factor
method. The analyzer includes the following standard options; precision frequency reference, internal
preamplifier, fine step attenuator, noise floor extension, 25 MHz analysis bandwidth, U7227A, C, or
F USB external preamplifier, and a convenient snap-on pouch to hold accessory items. Four model
numbers covering frequency ranges 3.6, 7, 26.5, and 40 GHz can be extended to 110 GHz with a
block downconverter. The series offers a low instrument noise figure, full-featured signal analyzer, and IQ
Analyzer (Basic) modes that extend beyond a dedicated noise figure analyzer.

Signal/spectrum analyzers
Adding an optional noise figure measurement application to a versatile signal or spectrum analyzer is
an economical way to add noise figure capabilities. The accuracy and frequency range of this solution
depends on which base instrument it is installed. Signal/spectrum analyzers use the Y-factor method to
measure noise figure. Preamplification, either external or internal, often improves accuracy.

Determine Your Optimal Noise Figure Measurement Method 104

Network analyzers
If you need noise figure measurements with the highest accuracy, choose Keysight’s PNA-X network
analyzer with the noise figure option. This solution is based on the cold-source technique, and it allows
S-parameter and noise figure measurements with a single connection to the DUT.

When selecting an instrument to meet your needs, it is first important to select one that will cover the
frequency range of your DUT.

MEASURE NOISE FIGURE ANYWHERE

https://www.keysight.com/see/radar-ew-component-test

Determine Your Optimal Noise Figure Measurement Method 105

Chapter 11
Select the Appropriate Phase Noise Test Solution
Phase noise is the random short-term phase fluctuation of a carrier signal which can limit the
performance of a Doppler radar system. Doppler radars determine the velocity of a target by
measuring the small shifts in frequency that return echoes have undergone. In actual systems,
however, the return signal is much more than just the target echo. The return includes a ‘clutter’
signal from the large, stationary earth as shown in Figure 11.1. If this clutter return is decorrelated
by the delay time difference, the phase noise from the local oscillator can spread the spectral
density of the earth return signals to mask the target signal, partially or completely. Thus, phase
noise on the local oscillator or in any component in the transmit or receive signal chains can set the
minimum signal level that must be returned by a target to detectable.

Doppler radar transceiver

Moving A
fT Target
Transmitter
Doppler
signal Decorrelated
fT+fD clutter noise
LO
fT
Doppler
Receiver signal
Stationary
fT Object
fT+fD f
Decorrelated clutter noise

Figure 11.1. Effect of carrier phase noise in a Doppler radar system.

Select the Appropriate Phase Noise Test Solution 106

Phase Noise Measurement Modes
There are four primary phase noise measurement modes, tailored to the needs of measuring the phase
noise contribution of four corresponding types of Devices Under Test (DUTs) with the following phase
noise measurement topologies:

Absolute 1-Port
DUT Absolute phase noise measurements for DUTs with one
Measurement output port, such as oscillators or signal generators
Receiver
with internal references, require only a phase noise
measurement receiver to measure the output of the DUT
since the stimulus source is internal.

Residual 2-Port Non-Frequency Translating

DUT Residual phase noise measurements of two-port
Measurement non-frequency translating devices, such as amplifiers,
Receiver require both a phase noise measurement receiver and a
stimulus source, the latter to drive the input of the DUT.
Both the receiver and the stimulus source operate at the
same frequency.

Residual 2-Port Frequency Translating

DUT Residual phase noise measurements of two-port
f1 f2 frequency translating devices, such as multipliers and
Measurement
Receiver dividers, require both a phase noise measurement
receiver and a stimulus source that operate at different
frequencies.

Residual 3-Port Frequency Translating

Residual phase noise measurements of three-port
DUT
f1 f2 frequency translating devices, such as mixers, also require
Measurement
Receiver both a phase noise measurement receiver and a stimulus
f3 source that operate at different frequencies, but also a
second stimulus source for the Local Oscillator (LO) that
operates at a third frequency.

Figure 11.2. Phase noise measurement topologies

Note that all three residual two-port measurement modes also apply to devices with multiple outputs, if
the outputs can be measured one at a time.

Select the Appropriate Phase Noise Test Solution 107

Phase Noise Measurement Methods
There are two primary methods of measuring phase noise:

a.) Direct spectrum method,

b.) Carrier removal method.

Direct spectrum method

By sampling the carrier of the DUT, the direct spectrum method as employed by Keysight’s vector signal
and network analyzers immediately provides amplitude and phase information by performing a complex
FFT of the digitized time record. Unlike legacy noise measurement applications found in scalar analyzers
that measured the power spectral density of a swept signal trace and presented both the amplitude
and phase modulation contributions of noise, modern vector analyzers can suppress the amplitude
modulation noise component and present only the phase noise component as shown in Figure 11.3.

The direct spectrum method is the most convenient and lowest cost solution since it is included with
signal and network analyzers and does not require extra equipment. But due to the presence of the DUT’s
carrier signal, the measurement sensitivity is limited by the phase noise of the analyzer, typically set by the
dynamic range of the digitizer.

Only report
phase noise
Suppress
amplitude
noise

ag
M
Phase
0 deg

Figure 11.3. Phase noise direct spectrum method by employing vector analyzer removal of amplitude noise.

Select the Appropriate Phase Noise Test Solution 108

Carrier removal method
For the most demanding low phase noise measurements, increased sensitivity is obtained by nulling
the carrier, using a double-balanced mixer, and combining the DUT signal with a reference signal of the
same frequency that is 90 degrees out of phase, in ‘quadrature’. Without the carrier, the phase noise
measurement receiver’s Low Noise Amplifier (LNA) gain can be increased to amplify the noise signal far
enough to overcome the noise floor limitation of the digitizer without clipping. The theoretical sensitivity
limitation for phase noise measurements then becomes the difference between the power of the DUT and
thermal phase noise floor of -177 dBm at 290 degrees K, minus the noise contribution of the stimulus
and reference sources and LNAs. The Carrier Removal Method is employed by Keysight’s Signal Source
Analyzers (SSAs) and Phase Noise Measurement Systems (PNTSs).

Absolute Phase Noise Measurements using the Carrier Removal Method

There are two ways to make absolute measurements using the carrier removal method, Frequency
Modulation (FM) discrimination where the quadrature signal is driven by the DUT through a 90-degree
delay line, and a Phase-Locked Loop (PLL) reference source locked 90 degrees out of phase to the DUT.
The frequency discriminator method is the most cost-effective, using the least equipment and being well
suited for measuring free-running oscillators with large excursions from a nominal frequency. But FM
Discrimination has sensitivity limitations, radar oscillators are amongst the lowest noise, and typically are
not drifting in a free-running mode, so FM discrimination will not be discussed here. We will focus instead
on the PLL reference source method that is employed by Keysight’s SSAs and PNTSs.

To achieve quadrature for the absolute PLL reference source phase noise measurement, a reference
source is locked to the DUT by a PLL that drives the output voltage of the double-balanced mixer to 0V.
The absolute PLL phase noise measurement block diagram utilizing the carrier removal method is shown
in Figure 11.4.

Power Power
Phase Detector

Phase noise Phase noise

Oscillator under Frequency Frequency

test (DUT) Carrier removed with
only phase noise left
Power at DC

Frequency
Reference signal
at same carrier
frequency as
DUT
Figure 11.4. PLL reference source is used to remove the carrier for an absolute phase noise measurement utilizing
the carrier removal method.

Select the Appropriate Phase Noise Test Solution 109

Residual Phase Noise Measurements Using the Carrier Removal Method
To achieve quadrature for the residual measurement case, a stimulus source output power is split into
two paths. One path goes straight to the LO input of the double-balanced mixer, and the other path goes
through the DUT, then a delay line, then a phase shifter to achieve a total of 90 degree phase shift before
being injected in to the RF input of the mixer to obtain quadrature and remove the carrier. The two-port
non-frequency translating phase noise measurement block diagram utilizing the carrier removal method is
shown in Figure 11.5.

SSB Cal Osc

cos(ωt + G(t)dut)
PNTS

G(t)dut 90° NFch

φ Det
RF In RF
DUT φ φ FFT
LO

Stimulus
LO In
sin (ωt)
sin (ωt)
1 channels with SSB Cal

Figure 11.5. Residual phase noise measurement.

Two Channel Cross-Correlation

To eliminate the noise contribution of the reference source, LNA, and receiver, two-channel cross-
correlation is employed by Keysight’s SSA and PNTS. The FFT of one channel is multiplied by the
complex conjugate of the other channel and then averaged over subsequent FFTs which maintains
the coherent DUT noise found on both channels but diminishes the non-coherent noise generated
independently by the equipment in each channel of the measurement system. 5 dB sensitivity
improvement is gained by every order of magnitude more cross-correlations. This adds some complexity
and can significantly add to measurement time if the reference sources have significantly higher phase
noise than the DUT, but it allows measurements made down to the thermal noise floor given enough time.

Select the Appropriate Phase Noise Test Solution 110

Table 11.1 shows system sensitivity vs. DUT input power.

Pcarrier, dBm Phase Noise L (fm), dBc/Hz

+20 -197
+10 -187
0 -177
-10 -167
-20 -157

Table 11.1. Theoretical two-channel cross-correlated sensitivity limits vs. DUT output power.

For faster measurement time of improved sensitivity measurements, Keysight’s PNTS provides the
additional flexibility to use high performance external references and stimulus sources, including
duplicates of one-port DUTs. Table 11.2 shows the improvement of sensitivity with number of cross-
correlations.

Number of cross-correlations, M 10 100 1,000 10,000

Noise reduction on N 1 + N2 -5 dB -10 dB -15 dB -20 dB

Table 11.2. Phase noise measurement sensitivity improvement above the thermal noise floor vs. number of
cross-correlations.

Keysight Phase Noise Measurement Solutions

Different solutions are available for measuring phase noise. The appropriate solution will depend on cost
and performance constraints. Although phase noise measurements on CW signals within the radar,
including the Stable Local Oscillator (STALO) and Coherent Oscillator (COHO), are critical, it may also be
necessary to measure the phase noise of pulsed signals or understand the phase noise contributed by
system components such as the power amplifier (residual or additive phase noise). This is especially true
for Doppler radars in which it is critical to understand the phase noise through the transmit path under
normal operating conditions.

A signal analyzer-based phase noise measurement solution typically offers the lowest cost option for
absolute measurement. Oscilloscopes can measure very wide (multi-GHz) offset on high-frequency
carriers. Modern VNAs incorporating high-quality signal sources can make absolute and residual
measurements at offsets to 10 MHz. A dedicated signal source analyzer such as the Keysight Signal
Source Analyzer (SSA) provides high performance and efficiency for measuring oscillators and phase
locked loops (PLLs) at offsets to 30 MHz. A Phase Noise Test System is more complex but offers the
greatest flexibility and performance including measurement of pulsed signals.

Select the Appropriate Phase Noise Test Solution 111

The choice of measurement solution will depend on DUT characteristics such as operating frequency,
measurement offset, and the target specification to be measured. Table 11.3 provides a comparison
of four phase noise measurement solution families used for radar and EW, followed by a brief overview
for each.

Attribute Infiniium N9068A Phase S930317B PNA E5052B Signal N5511A Phase
UXR, MXR Noise X-Series and PNA-X Source Analyzer Noise Test
Oscilloscopes Signal Analyzers Network System using 2
Measurement Analyzers E8257D Option
Application Phase Noise UNY PSGs
Measurement
Application
Absolute Yes Yes Yes Yes Yes
Measurements
Residual No Yes, but with Yes No Yes
Measurements external stimulus
source and
subtraction
Frequency DC to 110 GHz 3 Hz to 110 GHz; 10 MHz to 125 10 MHz to 26.5 50 kHz to 40
Range >>110 GHz with GHz GHz; GHz;
external mixer >>110 GHz with >> 110 GHz with
external mixer external mixer
Frequency 1 KHz to 55 GHz 1 Hz to 10 MHz 0.1 Hz to 10 1 Hz to 100 MHz 0.01 Hz to 160
Offsets MHz MHz
Noise Floor at 10 -121 dBc/Hz -134 dBc/Hz -140 dBc/Hz -137 dBc/Hz -170 dBc/Hz
GHz carrier and (after 1 cross (after 1 cross
10 kHz offset correlation at 1 correlation at 1
Hz offset) Hz offset)
Cross Yes No No Yes Yes
Correlation
VCO Yes (requires No No Yes Yes (FM
Measurements external DC Discriminator
source) Method)
Baseband Noise Yes No No Yes Yes
Measurements
AM Noise Combined Combined Combined Combined Separate Input
Measurement
Other Oscilloscope Signal Analyzer Network Signal Analyzer None
Measurements Analyzer (subset)
Measurement Moderate Low Low Low High (external
System (external (all internal) (all internal) (all internal) splitters and
Complexity splitters) signal sources)
Method Direct Direct Direct Carrier Carrier
Spectrum Spectrum Spectrum Removal Removal

Table 11.3. Keysight phase noise measurement solution portfolio comparison.

Select the Appropriate Phase Noise Test Solution 112

Infiniium real-time oscilloscopes
Keysight’s Infiniium real-time oscilloscopes now provide FFT phase noise measurements. To overcome
lower sensitivity that comes as a tradeoff for wider bandwidth digitization, two channels can now be used
to cross-correlate and reduce the noise contribution of the measurement receiver. This requires adding
a power splitter or double-probing each signal being measured to route it to two channels. Differential
signals can be measured directly with 4-channel oscilloscopes (there is no need for an external balun to
transform from balanced to unbalanced signal). Figure 11.6 shows the oscilloscope phase noise interface.

Figure 11.6. Infiniium Real-Time Oscilloscope phase noise interface.

X-series signal analyzer phase noise measurement application

Many spectrum analyzers include automated single-sideband (SSB) phase noise measurement functions
for CW signals. The range of offsets and levels of the phase noise that can be measured by a signal
analyzer will depend on the available RBW settings and the phase noise of the instrument itself.

Select the Appropriate Phase Noise Test Solution 113

The Keysight N9068A phase noise X-series signal analyzers measurement application can make absolute
phase noise measurements of one port devices and is available on all X-Series spectrum analyzers. With
the addition of a low phase noise stimulus source such as the Keysight E8257D Option UNY Precision
Signal Generator (PSG), it can also be used to make residual measurements on two-port devices by first
measuring the PSG, measuring the device under test (DUT), and subtracting the results. The N9068A
application interface is shown in Figure 11.7.

Figure 11.7. Keysight X-Series signal analyzers phase noise measurement application.

Network analyzer phase noise measurement application

With significant advancements made in the quality of signal sources embedded within network analyzers,
it becomes possible to make a phase noise measurement using a VNA. This has the convenience of
capturing multiple parametric measurements at a single test station with the DUT connected one time.

Keysight PNA Microwave Network Analyzers are fitted with enhanced low phase noise Direct Digital
Synthesis (DDS) stimulus sources. A dedicated phase noise measurement application supports absolute
phase noise measurements and residual noise measurements, without the need for external excitation
sources. The phase noise measurement sensitivity is -97 dBc/Hz1 at 100 Hz offset from a 10 GHz carrier.

Select the Appropriate Phase Noise Test Solution 114

The offset frequency range is 0.1 Hz to 10 MHz, and the sweep speed (typical) is 34 seconds (1 Hz to
10 MHz offset in normal mode). This application can also make AM noise, spurious and integrated noise
measurements.

Embedded-LO frequency converter Signal Source under Test

Figure 11.8. Residual and Absolute Phase Noise Measurement on a PNA.

The PNA is architected with multiple internal DDS sources (RF and LO sources) and multiple receivers.
The internal sources are all correlated. When you make 2-port or 3-port device measurements, the
sources cancel the phase noise and provide better measurement sensitivity, especially at close-in
offset frequencies. The sensitivity of frequency converter embedded-LO phase noise measurement is
determined by the phase noise of the DDS source at the carrier frequency of the embedded LO. When
the PNA provides both IF and LO or RF and LO a mixer under test, the phase noise cancellation is even
larger and makes the sensitivity drastically high, -86 dBc/Hz1 at 1 Hz offset from a 10 GHz carrier.

The N522xB PNA and N524xB PNA-X with S930317B/S930321B phase noise measurement application
simplifies your active device measurement configuration by using the same cable connection as other
measurements such as S-parameters, noise figure, gain compression, IMD, group delay, and reduces
the total test time significantly. Higher frequency models of the PNA like 67 GHz N5247B PNA-X and 125
GHz N5291A allow you to perform the phase noise measurement at millimeter-wave frequencies without
using harmonic mixers.

Select the Appropriate Phase Noise Test Solution 115

Figure 11.9. Keysight PNA phase noise application includes capability to measure AM noise and spurious.

Signal source analyzer

The Keysight E5052B signal source analyzer is one such instrument that employs cross-correlation
techniques to dramatically improve measurement performance. For convenience, the analyzer includes
its own reference oscillators. The SSA is designed to efficiently and rapidly process cross-correlations to
obtain maximum performance and efficiency. As a result, the SSA offers a very high level of performance
while keeping costs low.

Figure 11.10. Keysight signal source analyzer.

Select the Appropriate Phase Noise Test Solution 116

In addition to phase noise, the DSP-based SSA also provides many other functions that can be useful for
the testing of radar oscillators. Because it samples the signal and is DSP-based, the analyzer includes
the ability to analyze amplitude, frequency, and phase as a function of time. It can also perform transient
analysis by triggering frequency anomalies. The SSA provides two receiver channels: a wideband channel
to monitor frequency change and a narrowband channel to capture the frequency-versus-time profile
precisely. The signal can be measured simultaneously on both channels while either the wideband or
narrowband channel monitors the signal for frequency changes. These frequency changes can then
trigger and capture the transient event.

Phase noise test system

Dedicated test systems represent the “gold standard” for phase noise measurement. The Keysight
N5511A Phase Noise Test System (PNTS) can measure phase noise down to kT (-177 dBm/Hz at room
temperature). This thermal phase noise floor is the theoretical limit for any measurement. Therefore,
the PNTS can measure at the limits of physics. A modular instrument architecture takes advantage of
standalone instrumentation for a superior frequency offset range of 160MHz, broadest capability, best
sensitivity, and excellent speed of phase noise measurement.

Figure 11.11. Keysight phase noise test system with two PSGs.

Select the Appropriate Phase Noise Test Solution 117

The first architectural feature of the PNTS that minimizes the cross-correlation time penalty is the field
programmable gate array (FPGA) processing of all FFTs and cross-correlations in real-time. This means
that cross-spectral averaging is performed in FPGA hardware rather than in software on a microprocessor
and is therefore much faster. The computations are performed on all offset frequency segments of the
trace in parallel (multi-segment parallel processing) and displayed in real-time.

The second architectural feature that PNTS has that minimizes the cross-correlation time penalty is the
ability to use any electronic frequency tunable reference. This allows the use of the best available signal
generators today (or even use copies of the DUT as references) and gets better performance in the future
as newer and higher-performance signal generators come to market. This reduces the cross-correlation
time penalty because lower phase noise (better performance) references add less uncorrelated noise to
the measurement and thus require fewer correlations (and less time) to suppress this lower uncorrelated
reference noise and allow the user to observe their device’s true phase noise.

Pulsed and residual phase noise measurements are especially helpful for pulsed radar systems. The
N5511A can perform residual measurements on pulsed RF carriers as well as absolute measurements on
pulsed carriers. The phase detector modules in the N5511A come equipped with multiple internal PRF
(pulse repetition frequency) filters; these are low pass filters (LPFs) because all analog and digital signal
processing performed after the phase detectors in PNTS is performed at baseband. However, if there is
a measurement requirement for LPFs with different cutoff frequencies than the ones provided, N5511A
offers the flexibility of adding user-supplied external filters.

PHASE NOISE TEST SYSTEM

https://www.keysight.com/see/radar-ew-component-test

Select the Appropriate Phase Noise Test Solution 118

Chapter 12
Measure S-Parameters With Increased Confidence
Network analyzers characterize radio frequency (RF) devices. Although they began measuring
S-parameters, network analyzers have become highly integrated and advanced to stay ahead of
the devices they test. Today, most network analyzers are vector network analyzers — measuring
both magnitude and phase. Vector network analyzers are extremely versatile instruments that can
characterize scattering-parameters (S-Parameters), match complex impedances, make pulsed
measurements, measure distortion, noise figure, phase noise, and more.

Keysight offers the broadest range of network analyzer models and form factors — from
the portable FieldFox to the highly integrated PNA which can be configured with three high-
performance, low phase-noise signal generators.

Figure 12.1. Keysight network analyzer portfolio

Measure S-Parameters With Increased Confidence 119

S-Parameter Basics
S-Parameter design and measurement fundamentals is a topic beyond the scope of this chapter.

The number of S-parameters for a given device is equal to the square of the number of ports. For
example, a two-port device has four S-parameters. The numbering convention for S-parameters is that
the first number following the S is the port at which energy emerges, and the second number is the port
at which energy enters. So, S21 is a measure of power emerging from Port 2 as a result of applying an
RF stimulus to Port 1. When the numbers are the same (e.g. S11), a reflection measurement is indicated.

The Power of S-parameters

• Relate to familiar measurements (gain, loss, reflection coefficient, etc.)

• Relatively easy to measure
• Can cascade S-parameters of multiple devices to predict system performance
• Analytically convenient
- CAD programs
- Flow-graph analysis

Incident S 21 Transmitted
a
1 S 11 b2

Reflected DUT
S 22
b1 Port 1 Port 2 Reflected
a2
Transmitted S 12 Incident

b 1 = S 11 a 1 + S 12 a 2
b 2 = S 21 a 1 + S 22 a 2

Figure 12.2. S-parameters

Measure S-Parameters With Increased Confidence 120

Forward S-parameters are determined by measuring the magnitude and phase of the incident, reflected,
and transmitted signals when the output is terminated in a load that is precisely equal to the characteristic
impedance of the test system. In the case of a simple two-port network, S11 is equivalent to the input
complex reflection coefficient or impedance of the DUT, while S21 is the forward complex transmission
coefficient. By placing the source at the output port of the DUT and terminating the input port in a perfect
load, it is possible to measure the other two (reverse) S-parameters. Parameter S22 is equivalent to the
output complex reflection coefficient or output impedance of the DUT while S12 is the reverse complex
transmission coefficient (Figure 12.3).

S 21 b2
a1 Incident Transmitted
S 11 Z0
Forward
Reflected DUT load
b1 a2 = 0

Reflected b1 Reflected b2
S 11 = = a 2 =0 S 22 = = a1 =0
Incident a1 Incident a2
Transmitted b2 Transmitted b
S 21 = = a 2 =0 S 12 = = 1 a1 =0
Incident a1 Incident a2

a 1= 0
b2
Z0 S 22
DUT Reverse
load Reflected
a2
Transmitted S 12 Incident
b1

Figure 12.3. Measuring S-parameters

How to Boost and Attenuate Signal Levels When Measuring High-Power

Amplifiers
Testing high-power amplifiers can sometimes be challenging since the signal levels needed for the test
may be beyond the stimulus/response range of the network analyzer. High-power amplifiers often require
high input levels to characterize them under conditions similar to actual operation. Often these realistic
operating conditions also mean the output power of the amplifier exceeds the compression or burn-out
level of the analyzer’s receiver.

Measure S-Parameters With Increased Confidence 121

When you need an input level higher than the network analyzer’s source can provide, a preamplifier can
be used to boost the power level before the amplifier under test (AUT). By using a coupler on the output
of the preamplifier, a portion of the boosted input signal can be used for the analyzer’s reference channel
(Figure 12.4). This configuration removes the preamplifier’s frequency response and drift errors (by
rationing), which yields an accurate measurement of the AUT alone.

When the output power of the AUT exceeds the input compression level of the analyzer’s receiver,
some type of attenuation is needed to reduce the output level. This can be accomplished by using
couplers, attenuators, or a combination of both. Care must be taken to choose components that can
absorb the high power from the AUT without sustaining damage. Most loads designed for small-signal
use can only handle up to about one watt of power. Beyond that, special loads that can dissipate more
power must be used.

The frequency-response effects of the attenuators and couplers can be removed or minimized by using
the appropriate type of error-correction. One concern when calibrating with extra attenuation is that the
input levels to the receiver may be low during the calibration cycle. The power levels must be significantly
above the noise floor of the receiver for accurate measurements. For this reason, network analyzers that
have narrowband tuned receivers are typically used for high-power applications since their noise floor is
typically 90 dBm, and they exhibit excellent receiver linearity over a wide range of power levels.

Ref In

Coupler
AUT
Preamp High-power
load

Figure 12.4. High-powered forward measurement configuration

Measure S-Parameters With Increased Confidence 122

Some network analyzers with full two-port S-parameter capability enable measuring of the reverse
characteristics of the AUT to allow full two-port error correction. In this configuration, the preamplifier
must be added to the signal path before the port 1 coupler (Figure 12.5). Otherwise, the preamplifier’s
reverse isolation will prevent accurate measurements from being made on port 1. If attenuation is added
to the output port of the analyzer, it is best to use a higher power in the reverse direction to reduce noise
effects in the measurement of S22 and S12. Many VNAs allow uncoupling of the test port power to
accommodate different levels in the forward and reverse directions.

Network analyzer

RF amplifier Attenuator
Attenuator
"R"
Channel in

Coupler
Switch
Attenuator
DUT

Power
divider

Attenuator

Attenuator

Figure 12.5. High-powered forward and reverse measurement configuration

Pulsed RF Component Testing

Many RF and microwave amplifiers used in commercial and aerospace/defense applications require
testing using a pulsed-RF stimulus. This section focuses on new pulsed-RF S-parameter measurement
techniques that can be accomplished with microwave vector network analyzers.

The topic of pulsed-RF testing is often focused on measuring the pulses themselves. This is critical,
for example, in evaluating radar system performance and effectiveness. When measuring components,
however, the pulses are merely the stimulus, and the VNA measures the effect that the device under
test (DUT) has on the pulsed stimulus. Any non-ideal behavior of the pulses themselves is removed from
the measurement since the VNA performs ratioed measurements. This means that each S-parameter
measurement compares a measured reflection or transmission response with the incident signal,
providing ratioed magnitude and phase results. Figure 12.6, shows the configuration for measuring

Measure S-Parameters With Increased Confidence 123

forward S-parameters: the R receiver measures the incident signal, the A receiver measures the reflection
response, and the B receiver measures the transmission response. S11 is the complex ratio of the A and
R receivers, and S21 is the complex ratio of the B and R receivers.

Incident
DUT
Reflected Transmitted
Source

Signal
separation

Incident (R) Reflected (A) Transmitted (B)

Receiver/detector

Processor/display

Figure 12.6. Simplified vector network analyzer block diagram

Measure S-Parameters With Increased Confidence 124

There are three major types of pulsed-RF measurements as shown in Figure 12.7. The first two are
pulsed S-parameter measurements, where a single data point is acquired for each carrier frequency.
The data is displayed in the frequency domain with magnitude and/or phase of transmission and/or
reflection. Average pulse measurements do not attempt to position the data point at a specific position
within the pulse. For each carrier frequency, the displayed S-parameter represents the average value of
the pulse. Point-in-pulse measurements result from acquiring data only during a specified gate width and
position (delay) within the pulse. There are different ways to do this in hardware, depending on the type
of detection used, which will be covered later. Pulse profile measurements display the magnitude and/
or phase of the pulse versus time, instead of frequency. The data is acquired at uniformly spaced time
positions across the pulse while the carrier frequency is fixed at some desired frequency.

Magnitude and phase data

averaged over duration of pulse
data
point
Average
Pulse

Data acquired only during specified

gate width and position within pulse

Point-in-
Pulse
Data acquired at uniformly spaced
time positions across pulse

Pulse
Profile

Figure 12.7. Average, point-in-pulse, and pulse profile measurements

Measure S-Parameters With Increased Confidence 125

Pulsed-RF detection techniques Figure 12.8 shows an important measure of a pulsed RF signal and its
relationship between the time and frequency domain. When a signal is switched on and off in the time
domain (pulsed), the signal’s spectrum in the frequency domain has a sin(x)/x response. The width of the
lobes is inversely related to the pulse width (PW). This means that as the pulses get shorter in duration,
the spectral energy is spread across a wider bandwidth. The spacing between the various spectral
components is equal to the pulse repetition frequency (PRF). If the PRF is 10 kHz, then the spacing of
the spectral components is 10 kHz. In the time domain, the repetition of pulses is expressed as pulse
repetition interval (PRI) or pulse repetition period (PRP), which are two terms with the same meaning.

Another important measure of a pulsed RF signal is its duty cycle. This is the amount of time the pulse
is on, compared to the period of the pulses. A duty cycle of 1 (100%) would be a CW signal. A duty
cycle of 0.1 (10%) means that the pulse is on for one-tenth of the overall pulse period. For a fixed
pulse width, increasing the PRF will increase the duty cycle. For a fixed PRF, increasing the pulse width
increases the duty cycle. The duty cycle will become an important pulse parameter when we look at
narrowband detection.

Figure 12.8. Pulsed-RF network analysis terminologies

Measure S-Parameters With Increased Confidence 126

Wideband detection can be used when the majority of the pulsed-RF spectrum is within the bandwidth
of the receiver. In this case, the pulsed-RF signal will be demodulated in the instrument, producing
baseband pulses. With wideband detection, the analyzer is synchronized with the pulse stream, and
data acquisition only occurs when the pulse is in the “on” state. This means that a pulse trigger that
is synchronized to the PRF must be present; for this reason, this technique is also called synchronous
acquisition mode. The advantage of the wideband mode is that there is no loss in dynamic range when
the pulses have a low duty cycle (long time between pulses). The measurement might take longer, but
since the analyzer is always sampling when the pulse is on, the signal-to-noise ratio is essentially constant
versus duty cycle. The disadvantage of this technique is that there is a lower limit to measurable pulse
widths. As shown in Figure 12.9, as the pulse width becomes narrower, the spectral energy spreads
out—once enough of the energy is outside the bandwidth of the receiver, the instrument cannot detect
the pulses properly.

Figure 12.9. Pulse width and receiver bandwidth with wideband detection in time and frequency domain

Measure S-Parameters With Increased Confidence 127

Narrowband detection is used when most of the pulsed-RF spectrum is outside the bandwidth of the
receiver. In other words, the pulse width is narrower than the minimum data acquisition period with
the widest available receiver bandwidth. With this technique, the entire pulse spectrum is removed by
filtering except for the central frequency component, which represents the frequency of the RF carrier.
After filtering, the pulsed-RF signal appears as a sinusoid or CW signal. With narrowband detection, the
analyzer samples are not synchronized with the incoming pulses (therefore no pulse trigger is required), so
the technique is also called asynchronous acquisition mode. Usually, the PRF is high compared to the IF
bandwidth of the receiver, so the technique is also sometimes called the “high PRF” mode.

Keysight has developed a novel way of achieving narrowband detection using wider IF bandwidths than
normal, by using unique “spectral nulling” and “software gating” techniques. These techniques let the user
trade dynamic range for speed, with the result almost always yielding faster measurements than those
obtained by conventional filtering.

The advantage to narrowband detection is that there is no lower pulse-width limit, regardless of how
broad the pulse spectrum is, most of it is filtered away, leaving only the central spectral component.
The disadvantage to narrowband detection is that measurement dynamic range is a function of the duty
cycle. As the duty cycle of the pulses becomes smaller (longer time between pulses), the power of the
central spectral component becomes smaller, resulting in less signal-to-noise ratio as shown in Figure
12.10. Using this method, the measurement dynamic range decreases as the duty cycle decreases.
This phenomenon is often called “pulse desensitization” and can be expressed as 20*log (duty cycle) in
dB. Modern VNAs employs several unique features to minimize this effect, resulting in considerably less
degradation in dynamic range.

Figure 12.10. Duty cycle (time domain) versus signal-to-noise ratio of center spectrum (frequency domain)

Measure S-Parameters With Increased Confidence 128

In conclusion, testing with pulsed RF stimulus is very important for radar and electronic warfare
systems. Modern network analyzers utilize two different detection schemes, each with advantages and
disadvantages. Both wideband and narrowband detection can be used for point-in-pulse and pulse-
profile measurements. Pulse-to-pulse measurements require wideband detection. Although wideband
detection has traditionally been the more widely used technique, narrowband detection is a powerful
alternative for analyzing pulsed-RF signals. Keysight’s unique spectral nulling technique improves
measurement speed considerably by using wider IF bandwidths. For radar and EW applications,
narrowband detection offers a superior dynamic range and speed compared to past methods. With
narrowband detection, there is no lower limit to pulse widths. Keysight’s PNA vector network analyzers
offer built-in pulse modulators and pulse generators providing a fully integrated solution for millimeter-
wave pulsed-RF measurements.

Reduce Measurement Uncertainty

RF measurements are extremely sensitive. Test cables, connectors, and fixtures affect the measurement.
You want to characterize the DUT; not the DUT and the cables connecting it to the network analyzer.

By default, network analyzers consider the DUT everything beyond the test ports. This consideration
means the network analyzer’s reference plane is at the test ports. Everything beyond the reference plane
is included in the measurement.

The two most common methods of calibration are Thru, Reflect, Line (TRL), and Short, Open, Load, Thru
(SOLT). These methods are different combinations of impedance and transmission measurements used to
characterize the cables and fixtures for calibration.

These calibration techniques involve connecting standards with known properties to the measurement
setup in place of the DUT. The network analyzer can apply corrections for cables and connectors by
comparing what it measures to the values of the standards.

Traditionally, calibration is performed with mechanical standards. Operators would individually make
each connection and let the instrument take a measurement. Full two-port calibration requires seven
mechanical connections. This process is time-consuming and creates possibilities for user error.

Electronic calibration modules can replicate the different types of loads with just one connection.
Electronic calibration is fast, repeatable, and limits wear on connectors.

CalPod calibration refresh modules are in-situ devices that can remove the effects of environmental
variations in test cables, connectors, adaptors, and switch matrices to re-establish a valid calibration at
the measurement plane, assuring your device’s performance is not affected by these other environmental
variations. Easily refresh a calibration at the push of a button, without removing the DUT, and without the
physical connection of standards.

Measure S-Parameters With Increased Confidence 129

De-Embedding Fixture Effects
Many of today’s devices do not have coaxial connectors and are put in fixtures to measure them in a
coaxial environment. Accurately removing the effects of the fixture is required to get a good measurement
of the DUT. Complicated modeling in EM simulation software or multiple calibration standards fabricated
on board were needed, adding complexity and expense.

Powerful, fast, and accurate automatic fixture removal (AFR) can handle a variety of measurement needs
such as:

• Single-ended and differential devices

• The left and right side of the fixture can be asymmetrical
• Through lengths can be specified or determined from open or short measurements
• Band-pass time-domain mode for band-limited devices
• Extrapolation to match DUT frequency range
• Power correction compensates for fixture loss versus frequency
• De-embed files can be saved in a variety of formats for later use in PNA, ADS, and PLTS

Keysight AFR is the fastest way to de-embed a fixture from the measurement.

Figure 12.11. The S93007B five-step wizard guides you through the process to characterize your fixture and
removes it from your measurement.

Measure S-Parameters With Increased Confidence 130

Simplifying Multiport Measurements
Modern devices are highly integrated requiring test and validation of devices with more than four ports.
Radar and EW subsystems such as RF transmit receive modules (TRM) and phased array antennas
require multiport characterization for all components. The need for multiport tests accelerated the
development of switch-based solutions for traditional vector network analyzers (VNAs). When switch-
based solutions were not adequate to keep up with the multiport test, the VNA itself was re-imagined and
optimized for the multiport test.

A B
2-port VNA

Port 1 Port 2

24-port simple
switch-tree test set

Figure 12.12. VNA with simple switch test set

High-volume tuning and testing of multiport devices can be greatly simplified by using a multiport network
analyzer, or a multiport test set with a traditional two-port analyzer.

A single connection to each port of the device under test (DUT) allows for complete testing of all
transmission paths and port reflection characteristics. Multiport test systems eliminate time-consuming
reconnections to the DUT, significantly increasing efficiency. Furthermore, the risk of misconnections
is lowered, operator fatigue is reduced, and the wear on cables, fixtures, connectors, and the DUT
is minimized.

Measure S-Parameters With Increased Confidence 131

Full crossbar switching test sets measure between every port of a device as well as apply a load
termination to unused ports. The full crossbar setup offers complete measurements between each of the
ports but also introduces new challenges. Each of the six test ports in Figure 12.13 has two possible 1x6
switches to use as a load termination. The switch used for termination on each port varies depending on
which ports are active. The varying termination of the ports makes full N-by-N calibration difficult because
you need to calibrate for every case.

A B

2-port VNA

Port 1 Port 2

6-port crossbar testset

Port 1 Port 2 Port 3 Port 4 Port 5 Port 6

Figure 12.13. Full crossbar test set

External switches significantly affect measurement performance. The switching hardware is beyond the
VNA’s directional couplers that sample the test signal. Interference from the switching hardware creates a
mismatch between the signal reaching the DUT and what the VNA measures. Calibration can help, but a
switch matrix can never achieve the dynamic range, temperature stability, and trace noise performance of
a standalone VNA.

Switch matrices provide a low-cost solution for multiport testing, but require significant operator
intervention for setup, calibration, and adjusting configurations. Modern multiport devices demand faster
and more accurate measurements than a switch-based solution.

True multiport setups maintain their measurement performance no matter how many ports you use. As
frequencies trend higher and margins for error shrink, your test setup’s measurement performance can
significantly impact your yield.

Removing switches from multiport measurements does more than speed up the measurements. Setups
without switches provide full VNA measurement performance. Attenuation from the switches degrades
the dynamic range of switch-based multiport measurements, particularly at higher frequencies. By 20
GHz, switch-based solutions retain only half the dynamic range of multiport solutions.

Measure S-Parameters With Increased Confidence 132

The multiport VNA makes sweeps quickly due to the superior dynamic range of multiport VNAs. The
attenuation of the switches in a switch-based system degrades dynamic range, especially at higher
frequencies. The consistently wide dynamic range of a multiport VNA means you can use a wider IF
bandwidth and capture your sweeps faster. Making fast measurements at higher frequencies becomes
crucial to keep up with the trend of higher data rates in both high-speed digital and RF.

Multiport setups require significant calibration time. Switched test sets with solid-state switches are
easily affected by temperature and must undergo frequent calibration to ensure accurate measurements.
Calibration downtime drastically impacts throughput; a system with longer calibration intervals reduces
your test time and your cost of test.

24
Number of ports on DUT

16 True Multiport PXI VNA

4-port VNA w/Extension TS

4-port VNA w/Full Crossbar TS

2-port VNA w/Full Crossbar TS

0 100 200 300 400 500

Number of sweeps for an N-port measurement
Figure 12.14. Sweeps required for multiport devices

Measure S-Parameters With Increased Confidence 133

A multiport VNA significantly reduces your test time to give you highly accurate measurements on
multiport devices. The flexible modular configuration, fast measurement times, and high performance
make VNAs the right tool for multiport testing. VNAs with up to 50 ports on a single chassis has the
flexibility and performance to take on the toughest multiport challenges.

DEVICE EXPERT: GUIDED MEASUREMENT SETUP ON

THE PNA NETWORK ANALYZER REAL TIME UNCERTAINTY AT MILLIMETER WAVE FREQUENCIES

DC-67 GHZ ELECTRONIC VNA CALIBRATION

https://www.keysight.com/see/radar-ew-component-test

Measure S-Parameters With Increased Confidence 134

Chapter 13
Characterizing Frequency Translating Devices
Frequency-translating devices like upconverters and downconverters are one of the fundamental
components in every RF or microwave transceiver chain and play fundamental roles in the transmit
and receive chains of all defense systems, ranging from EW, ECM, ESM, ELINT, and SIGINT
receivers, to satellite terminals and transponders, and radar systems. Radar, EW, and satellite
RF systems require frequency converters or mixers with specified and well-controlled amplitude
and group delay response. These key components present unique measurement challenges
because they exhibit both desired and undesired linear and non-linear behavior. There are several
measurements needed to fully characterize frequency-converting devices.

• Transmission measurements include RF to IF transfer function viz. conversion loss or gain, group
delay, derivation from linear phase, and port to port isolation.
• Reflection measurements including return loss and VSWR.
• Measurements that characterize the distortion added by the conversion process include
intermodulation distortion, conversion compression, and undesired mixing products.

A radio receiving system requires that the mixers within it have well-controlled amplitude, phase,
and group-delay responses. Due to the nature of the required measurements, a stimulus-response
ratioed measurement technique, such as S-parameter analysis, is ideal for most of the above
measurements. Unfortunately, since these components have different input and output frequencies
traditional S-parameter and network analysis techniques used for non-frequency translating
devices are not valid.

Characterizing Frequency Translating Devices 135

Test Methodology
Vector network analyzers are an important measurement tool for characterizing the magnitude, phase,
group delay, impedance, linearity, and isolation performance of high-frequency components. To
accomplish this, a network analyzer provides a stimulus source, signal-separation devices, receivers
for signal detection, display/processing circuitry, and algorithms for reviewing results. The hardware
architecture of the PNA Series network analyzer as shown in Figure 13.1 uses narrowband mixer-based
receivers. However, since the output is different from the input frequency for frequency converting
devices, this tuned-receiver technique needs to be modified. To make these measurements possible the
receivers must be independently tuned to a fixed offset from the stimulus.

Reference Measurement
receiver receiver

R2 B
Splitter

Signal separation
Source LO

R1 A
Reference Measurement
receiver receiver

Figure 13.1. General network analyzer block diagram

This technique is referred to as frequency offset and is an option on many vector network analyzers.
This requires additional hardware and firmware control. Traditional techniques used for calibration and
error correction are not valid with frequency-offset hardware changes because the error terms cannot be
maintained in a constant phase relation. A modified frequency-offset error model is needed to describe
the interactions of error signals in this environment. Frequency Offset Mode (FOM) provides the capability
to have the VNA Sources tuned to frequencies that are different (offset) from the VNA Receivers. Further,
modern mixers and frequency converters also push the limits of a network analyzer’s capabilities. Getting
consistent measurements for operating bandwidths as wide as 3.2 GHz at 29 GHz often requires
averaging multiple sweeps to reduce the noise.

Characterizing Frequency Translating Devices 136

To meet the demands of modern high-frequency device testing, Keysight introduced in its PNA and
PNA-X Series of network analyzers to use the same direct-digital synthesis (DDS) technology as high-end
signal generators. DDS signal sources provide consistent and clean signals for unmatched performance in
sensitive applications. The new sources with DDS with very low phase noise make phase measurements
more stable and improves the measurement performance of some applications that require stable phase
noise measurements: scalar-mixer calibration (SMC) + phase.

The frequency converter application (FCA) is an option offered with the microwave PNA Series network
analyzers and is designed to address both the calibration and measurement difficulties for testing
frequency converter devices. FCA offers an easy-to-use graphical user interface and advanced calibration
techniques, including the scalar-mixer calibration, scalar-mixer + phase (SMC+Phase), and vector-mixer
calibration (VMC).

SMC can be used to characterize the conversion loss magnitude and reflection parameters of mixers. A
conversion loss magnitude measurement is a ratio of the output power (at the output frequency) to the
input power (at the input frequency). The calibration is based on a combination of port and device match
characterization and power meter measurements. With SMC, the input and output power levels are
accurately determined by calibrating the network analyzer with a power meter, thereby transferring the
accuracy of a power meter to the network analyzer. By using the network analyzer’s one-port calibration
ability, the port and device input and output reflection coefficients are measured. Using the known vector
reflection coefficients of the test port, the device, and the power sensor, SMC corrects for mismatch loss.
Since SMC is referenced to a traceable standard (power sensor/meter measurements), it provides the
best-specified measurement of conversion loss magnitude.

VMC offers measurements of conversion loss magnitude, phase, and absolute group delay by using a
combination of calibration standards (such as short, open, load, or ECal), and a “calibration mixer/IF-filter”
pair during calibration. Vector-mixer calibration is based on a modified two-port error model, however,
the steps and standards used to determine the error terms differ from the traditional two-port calibration.
In the case of a frequency-translating device, the procedures differ because the input and output
frequencies are different and additional calibration steps are required. In VMC, the calibration standards
are still used to determine the directivity and match error terms. A “calibration mixer/IF-filter” pair is
used as a new standard to determine the transmission tracking term. A calibration mixer that is assumed
to be reciprocal, is characterized for input match, output match, and conversion loss (both magnitude
and phase).

SMC+Phase combines the simplicity of SMC with the phase and delay measurement capability of
VMC and eliminates the need for reference and calibration mixers for phase or group delay testing.
SMC+Phase can also be used to measure converters with embedded LOs that are difficult or impossible
to access. This technique significantly simplifies and reduces the cost of the measurement test setup.

Characterizing Frequency Translating Devices 137

As explained earlier, SMC is the most accurate way to measure conversion loss and gain. It corrects
for mismatch errors during calibration and measurements by combining one port and power-meter
calibrations. The technique is simple to set up and calibrate and requires a power meter during calibration
along with the usual open, short, load, and thru standards. SMC corrects for DUT mismatch during
transmission measurements by taking advantage of the VNA’s ability to measure its source and load
match during calibration as well as the DUT’s input and output match. Two techniques are shown in
Figure 13.2.

Power
sensor
Calibration
VMC mixer/filter
SMC
Reference
mixer
ECal
DUT module DUT

Figure 13.2. SMC + VMC for frequency translating devices

The VMC technique delivers the most accurate phase and absolute group delay measurements,
calibrates the test system’s transmission phase response, and provides mismatch correction at the input
and output of the DUT. However, the technique has several inherent drawbacks. It is more complicated
and requires more external components than SMC as two additional mixers are needed for reference
and calibration. VMC uses a characterized mixer as a calibration thru standard along with the usual
open, short, and load standards, and removes magnitude and phase errors for transmission and
reflection measurements. An external reference mixer is used as a phase reference but is not needed for
phase locking the source and receivers with a frequency offset, as offset sweeps are achieved with the
instrument’s internal sources. As both calibration techniques perform corrections for mismatches, external
attenuators are rarely needed. While VMC provides the ability to evaluate deviation from linear phase and
absolute group delay, mixers that match the frequency range of the DUT are harder to obtain above 26.5
GHz as are filters with acceptable performance. In addition, many mixers may be required to evaluate
DUTs with diverse frequency plans, so several calibrations must often be performed to cover all bands.

The SMC+Phase technique, as shown in Figure 13.3, uses simple setup and calibration, requires no
external signal source or reference and calibration mixers, provides the most accurate conversion-loss/
gain and phase/delay measurements, and removes mismatch errors during calibration and measurement.
It replaces VMC for most frequency converter measurement applications. In contrast to VMC that
uses ratios of test and reference signals at the same frequency (thus the required reference mixer), the
SMC+Phase technique ratios single-receiver phase measurements performed at the DUT’s input and
output. It also replaces the calibration mixer with a comb generator as a phase standard, and magnitude

Characterizing Frequency Translating Devices 138

measurements are performed the same way as SMC, using a power sensor as a calibration standard.
By eliminating the reference and calibration mixers, SMC+Phase simplifies and reduces the cost of the
converter measurement system.

Power
10 MHz sensor

Comb
Gen

SMC+Phase ECal
DUT module

Figure 13.3. Setup for SMC+Phase

The phase-measurement technique employed by SMC+Phase relies on the phase coherency of the
signal sources in the instruments’ DDS architecture to eliminate the reference mixer Figure 13.4. This is
an advantage of the PNA and PNA-X over other VNAs that do not have the new DDS sources. The PNA
and PNA-X feature extremely low phase noise and spurious emissions to help you measure complex
components faster. DDS sources maintain relative phase coherence across a frequency sweep by
digitally incrementing the phase accumulators and by employing synchronous IF detection and digital
signal processing.

Frequency Numerically Reconstruction

Analog
control controlled DAC lowpass
output
register oscillator filter

Reference oscillator Fclk

Figure 13.4. Direct Digital Synthesis

Characterizing Frequency Translating Devices 139

In the first step of this new method, the VNA receivers are calibrated for absolute power and phase
relative to the test ports without test cables or other system interconnects attached. The second step
is a simple S-parameter calibration that removes the effects of test cables, adapters, attenuators, and
wafer probes. The comb generator creates a repetitive, negative-going impulse in the time domain, which
provides a broadband phase-calibrated frequency spectrum (i.e., a “comb” of signals) that is used to
calibrate the phase of the VNA receivers (Figure 13.5).

Figure 13.5. Comb Generator

The calibration performed during the first step adjusts the VNA receivers for absolute power using a
power sensor and phase using a comb generator. It is typically performed directly at the reference
plane of the instrument’s test ports or with adapters connected to them. This eliminates the effect of
test cables, making the calibration more accurate and repeatable. Calibration is typically performed over
the full frequency range of the instrument and can be performed infrequently because of the
instruments’ stability.

The S-parameter calibration performed in the second step includes the system interconnect hardware
that was not included in the first step and is performed at the end of the test cables, adapters, or wafer
probes that connect to the DUT. For coaxial calibrations in which the ECal module connectors match
those of the DUT, this calibration can be done as a single step.

Characterizing Frequency Translating Devices 140

Advanced Parameters Test Methodology
Embedded LOs are common in certain types of converters and transponders. As a result, it is impossible
to access the DUT’s LO or its time base and thus also impossible to make the necessary connections
to provide coherent frequency synchronization of the VNA and the transponder (Figure 13.6). Therefore,
VNAs have traditionally not been used for these measurements, which has hindered efforts to reduce
the time required for transponder characterization. Keysight has created a way to circumvent this
problem that makes it possible to make very accurate VNA-based converter measurements in these
situations. There are four basic obstacles that must be surmounted to make VNA-based measurements
of embedded LOs possible. The first is frequency stability, which is not an issue for most transponders, as
their LOs are locked to highly stable crystal oscillators and have low phase noise.

10 MHz

DUT

PROBLEM
• Can’t lock internal LO of DUT to VNA
• VNA receivers won’t be tuned to
exact output frequency resulting
in unstable phase/delay measurements

Figure 13.6. Embedded LOs, a fact of life

Characterizing Frequency Translating Devices 141

The second issue is establishing frequency coherence because the narrowband VNA receivers must be
tuned to frequencies that exactly match the output frequencies of the DUT. The VNA’s LO must also be
stable enough relative to the DUT’s LO to ensure that errors caused by the non-ratioed phase are not
excessive. The next issue is phase stability that arises because even when the frequencies are the same
there will be sweep-to sweep variations in absolute phase response caused by the architecture of the
instrument. Fortunately, phase can be normalized at each sweep to an arbitrary phase reference so that
averaging can be used just as effectively as with a common time-base.

Finally, as the phase noise of the LOs in the VNA and DUT cannot be “ratioed out”, averaging and
smoothing can be used to lower the noise of the group delay measurements. To establish the appropriate
pseudo-coherent phase relationship between the DUT and the test instrument, the PNA or PNA-X breaks
down the measurement of the converter’s effective LO into coarse and fine measurements. This two-step
approach quickly achieves the needed frequency accuracy. Coarse tuning is achieved by first setting
the RF stimulus to an appropriate CW frequency within the defined input frequency band, and then the
instrument calculates the output frequency corresponding to the input frequency and the nominal value of
the DUT’s LO.

The instrument’s internal receivers are swept around the expected center frequency of the DUT’s output.
The difference between the peak of the actual signal and the expected signal (based on the nominal value
of the DUT’s LO) gives a frequency-offset value for adjusting the nominal LO value of the mixing plan. The
VNA is then tuned very close to the DUT’s actual output frequency. The frequency span of the course
receiver sweep can be set by the user up to 10 MHz. The coarse-tune process gets the instrument close
to the desired output frequency but not close enough to stop phase slippage between the VNA and the
DUT. The necessary frequency accuracy can be obtained by taking a different measurement approach
for the fine sweep. Once the coarse offset is applied, the PNA performs a ratioed phase-versus time
sweep between the DUT and test receivers at a fixed input frequency and with the test receiver fix-tuned
to the output frequency of the DUT derived with the coarse sweep. Any small residual frequency offset
will show up as a linear phase change versus time. The slope of this phase change can be accurately
estimated, which gives the fine offset value. This fine-tuning process can be repeated multiple times to
get a good sub-hertz estimate of the DUT’s LO frequency. Minimizing the phase shift versus time until the
phase response has a flat slope over the measurement period provides a pseudo-locked condition and
places the two local oscillators in a fixed phase relationship. This method is much faster than performing
a narrowband sweep of the VNA’s receivers with many data points. Both coarse and fine-tuning can be
performed at every sweep of the group delay measurement, creating a coherent relationship between the
instrument and the DUT.

Characterizing Frequency Translating Devices 142

Apart from the frequency response of the frequency translating devices, they need to be characterized for
other advanced parameters too, like IMD, compression, noise figure, modulation distortion, etc. Complex
measurements may require multiple instruments, but the 3rd source eliminates the need for an external
signal generator for mixer IMD measurements (Figure 13.7).

3rd Source

IMD IMD
Input Output

Figure 13.7. Mixer IMD measurement

FCA applications on the VNA provide the scalar mixer/converter plus phase (SMC+Phase) measurement
class that provides fully calibrated conversion gain/loss, relative phase, and absolute group delay
measurements of mixers and converters without the need for reference or calibration mixers.
Eliminating the calibration mixer requires a U9391C/F/G comb generator. FCA provides an intuitive
and easy-to-use user interface for setting up mixer and converter measurements, with single or dual
conversion stages. It can control the analyzer’s built-in source(s) as well as external signal generators
for use as LO signals. Supported external sources include the Keysight sources as well as other SCPI-
controlled signal generators.

Characterizing Frequency Translating Devices 143

VNA firmware supports numerous measurement classes for the complete characterization of frequency
translating devices. (Figure 13.8).

Figure 13.8. Measurement classes on PNA series VNA

Further, when measuring the noise figure of frequency converters, source power calibration is performed
which levels the source power versus frequency. The leveled power is later used to calibrate the standard
receivers for absolute power measurements, which is necessary for measuring the conversion gain (or
loss) of frequency converters. This methodology is the same as is used in the SMC measurement class.

Keysight ENA, PXIe, and USB streamline VNA also offer advanced mixer calibration techniques: SMC,
SMC+Phase, and vector-mixer calibration to test frequency translating devices.

Together these techniques represent significant advances in converter characterization that are likely to
become the standard for the way these measurements are made by the radar and EW system designers.

Characterizing Frequency Translating Devices 144

Chapter 14
Expand Insight into Nonlinear Behavior and Modeling
When designing systems with high-power amplifiers, designers measured S-parameters using
a vector network analyzer, loaded the results into an RF simulator, added other measured or
modeled circuit elements, and then ran a simulation to predict system performance such as gain
and power-efficiency under various loads. Since S-parameters assume that all elements in the
system are linear, this approach does not work well when attempting to simulate performance
when the amplifier is in compression or saturation, as real-world amplifiers often are. The errors
are particularly apparent when simulating the combined performance of two cascaded devices
that exhibit nonlinear behavior. While engineers may live with this inaccuracy, it invariably results in
extensive and costly empirical-based iterations of the design, adding substantial time and cost to
the design and verification process.

Expand Insight into Nonlinear Behavior and Modeling 145

Testing today’s high-power devices demands an alternate solution—one that quickly and accurately
measures and displays the device’s nonlinear behavior under large-signal conditions and provides an
accurate behavioral model that can be used for linear and nonlinear circuit simulations. Nonlinear vector
network analyzer (NVNA) and X-parameters provide a means to efficiently analyze and design active
devices and systems under real-world operating conditions, reduce design cycles by as much as 50%,
and providing valuable insight into device behavior with full nonlinear component characterization.

Amplifier characteristics depend on the impedance matching and need to be tested under actual drive
power. The power delivered and the optimum load can be characterized with normal S22 S-parameter
measurements in linear regions, but once the amplifier operates at high power, it goes into nonlinear
regions, and the behavior cannot be predicted.

Active hot parameters are appropriate for amplifiers where the transistors are pre-matched and are used
to verify that the matching is good and there are not any extraneous matching issues. The technique
is significantly faster than other methods and provides a measure of the true Hot S22, the optimum
match for maximum power, the value of the maximum power as well as the power delivered to 50 Ω.
In summary, it provides the fundamental X-parameters of the amplifier. This capability can be added to
PNA-X network analyzers with the S93110B Active Hot Parameters application.

Bare transistors, which require substantial impedance matching at fundamental and harmonic frequencies
require testing with the nonlinear vector network analyzer and X-parameters - the mathematically correct
extension of S-parameters to large-signal conditions. This provides a device-independent black-box
framework whose coefficients are identifiable from a simple set of physical measurements on the device
under test.

X-parameters are a fully nonlinear framework that provides both the magnitude and phase of the
fundamental and harmonics. They can be cascaded in simulation and produce the correct behavior in
mismatched environments. Researchers and designers can now measure match, gain, group delay, and
more for driven components. The methodology can be extended to two-tone, multi-tone, frequency
converter, and three-port mixer parameters.

The standard PNA-X is transformed into the NVNA with a minimum of external accessories and nonlinear
firmware options. Core to this transformation is the nonlinear calibration process. Trust in the measured
data is as important as the data itself. NVNA’s state-of-the-art nonlinear calibration process provides
vector calibrated amplitude and phase data traceable to the National Institute of Science and Technology
(NIST). A simple three-step calibration process is driven by a graphical calibration wizard to remove any
systematic errors and maximize accuracy.

Expand Insight into Nonlinear Behavior and Modeling 146

Used in conjunction with NVNA nonlinear measurements

PathWave design and simulation

tools, X-parameters minimize
design iterations, speed
simulation, and deterministically
model the nonlinear behavior of
your active components. This
can significantly reduce the
time to market for component,
module, and system design.
Additionally, because
Keysight’s X-parameters
ADS simulation and design
are a measurement-based
representation of the DUT, they
X-parameter X-parameter

v1 v2
1 0 1 0 1 0
I_Probe U_Probe
R
can be used to distribute more i1 i2
R11 MCA_ZFL_11AD Connector MCA_ZX60_2522
R=50 Ohm DC_Block MCA_ZFL_11AD_1 X1 MCA_ZX60_2522_1
DC_Block1 fundamental_1=fundamental fundamental_1=fundamental

complete device operating + V_1Tone

SCR14
V = polar(2*A1 1N,O) V
-
characteristics than traditional Freq-fundamental

X-parameter blocks
datasheets, and at the same
X-parameters enable accurate nonlinear
time protect the device IP. simulation under arbitrary matching conditions.

This allows prediction of component behavior

in complicated nonlinear circuits.

Figure 14.1. Measure complete linear and nonlinear component behavior

with the Keysight NVNA, and then accurately perform simulations and
optimizations with Keysight’s Advanced Design System.

COMPLETE LINEAR AND NON-LINEAR CHARACTERIZATION OF

ACTIVE COMPONENTS

https://www.keysight.com/see/radar-ew-component-test

Expand Insight into Nonlinear Behavior and Modeling 147

Chapter 15
Methods for Measuring the Dielectric Properties of Materials
Dielectric property is one of the fundamental electromagnetic properties of materials. It describes
how a material interacts with an applied electromagnetic field. When electric and magnetic fields
pass through a material, each can interact with that material in two ways:

• Storage: Energy may be exchanged between the field and the material, in a bi-directional
(lossless) manner.
• Loss: Energy may be permanently lost from the field, and absorbed in the material
(usually as heat).

Methods for Measuring the Dielectric Properties of Materials 148

Every material has a unique set of electrical characteristics that are dependent on its dielectric properties.
“Materials” can mean just about anything. They are produced or used by many diverse industries. For
example, the loss of a cable insulator, the impedance of a substrate for a microwave integrated circuit, or
the frequency of a dielectric resonator can be related to its dielectric properties. The information is also
useful for improving ferrite, absorber, and packaging designs.

Material Characterization
Material has a unique set of electrical and magnetic characteristics that are dependent on its
electromagnetic properties. Accurate measurements of these properties can provide scientists and
engineers with valuable information to properly incorporate the material into its intended application
for more solid designs or to monitor a manufacturing process for improved quality control.
A dielectric materials measurement can provide critical design parameter information for many
electronics applications.

Numerous techniques are adopted for material characterization. Here’s an overview of the techniques we
will be discussing in this chapter.

1. Parallel Plate (sometimes called capacitance method): It uses a parallel plate capacitor, with
the material in between. This method uses an impedance analyzer. It is typically used at the lower
frequencies, below 1 GHz.
2. Coaxial probe: This method uses an open-ended coaxial probe, usually with a network analyzer.
It is the easiest method to use for liquids, or soft semi-solids, although very flat hard solids can be
measured as well. Keysight offers probes in the RF to microwave frequencies, 200MHz to 50GHz.
3. Resonant Cavity: This method uses a resonant cavity for the sample holder, and a network analyzer
to measure the resonant frequency and Q of the cavity, both empty and with the sample present.
From this, permittivity can be calculated. This method has the best loss factor resolution.
4. Transmission Line: This method can use a variety of transmission “lines” for sample holders
with a network analyzer. Lines can be coaxial; waveguide and even free space are considered a
transmission line technique. It is useful for a broad frequency range, from the low microwave region to
mm-wave.

Each method has strengths and limitations that make it useful for a particular application. The choice
of technique depends on many things like frequency of interest, required measurement accuracy,
material properties (i.e., homogeneous, isotropic), the form of material (i.e., liquid, powder, solid, sheet);
sample size restrictions, destructive or nondestructive testing, contacting or non-contacting test,
measurement temperature.

Methods for Measuring the Dielectric Properties of Materials 149

Figure 15.1 provides a map of different measurement techniques according to the frequency range of
operation and the MUT (Material Under Test) losses.

Loss
Coaxial Probe
High

Transmission line

Medium Free Space

Parallel
Plate

Low Resonant Cavity

Frequency
Low 50 MHz 5 GHz 20 GHz 40 GHz
Microwave 60 GHz 325+GHz
frequency RI Millimeter-wave
Figure 15.1. Material measurement techniques

Parallel Plate Technique

The parallel plate method also called the three terminals method in ASTM standard D15012, involves
sandwiching a thin sheet of material or liquid between two electrodes to form a capacitor. The measured
capacitance is then used to calculate permittivity, see Figure 15.2. In an actual test setup, two electrodes
are configured with a test fixture sandwiching dielectric material. The impedance-measuring instrument
would measure vector components of capacitance (C) and dissipation (D) and a software program would
calculate permittivity and loss tangent.

Electrodes (area = A) Y = G + j ω Cp
Cp G
Equivalent circuit = jωC 0 –j
C0 ω C0

Co: Air capacitance

Cp G
Cp G
εr* = –j ω
C0 C0

t • Cp
εr' =
A • ε0

t
εr" =
Solid thickness = t Liquid ω • Rp • A • ε0

Figure 15.2. Parallel plate technique

Methods for Measuring the Dielectric Properties of Materials 150

The method works best for accurate, low-frequency measurements of thin sheets or liquids. A typical
measurement system using the parallel plate method consists of an impedance analyzer or LCR meter
and a fixture such as the 16451B and 16453A dielectric test fixture, which operates up to 1 GHz. The
16452A test fixture is offered for measuring liquids.

RF parallel plate provides ease of use and good accuracy for both dielectric and magnetic materials.
However, the parallel plate method does not measure materials with magnetic properties.

16454A

In
h
where,
relative permeability
measured inductance with MUT
c measured inductance without MUT
permeability of free space
b height of MUT (Material Under Test)
outer diameter of MUT
No magnetic flux leakage inner diameter of MUT

Figure 15.3. Measurement method for magnetic material

You should use the “inductance” measurement method to measure permeability. Relative permeability
of magnetic material derived from the self-inductance of a cored inductor that has a closed-loop (such
as the toroidal core) is often called effective permeability, see Figure 15.3. The conventional method
of measuring effective permeability is to wind some wire around the core and evaluate the inductance
with respect to the ends of the wire. This type of measurement is usually performed with an impedance
analyzer. Effective permeability is derived from the inductance measurement result. The Keysight 16454A
magnetic material test fixture provides an ideal structure for a single-turn inductor, with no flux leakage
when a toroidal core is inserted in it.

Coaxial Probe Method

A typical coaxial probe system consists of a vector network analyzer, a coaxial probe, and software to
calculate permittivity from calibrated S-parameter measurements. Keysight offers probes and software for
these measurements.

The coaxial probe method works with Keysight N1501A dielectric probe hardware. Measurements are
conveniently made by immersing the probe into liquids or semi-solids – no special fixtures or containers
are required. Measurements are non-destructive and can be made in real-time. These important features
allow the dielectric probe kit to be used in process analytic technologies. The open-ended coaxial probe

Methods for Measuring the Dielectric Properties of Materials 151

is a cut-off section of the transmission line. The material is measured by touching the probe to a flat face
of a solid or immersing it into a liquid or semisolid. The fields at the probe end “fringe” into the material
and change as they meet the MUT. The reflected signal (S11) can be measured and related to εr.

Reflection
(S11 )
15.4. Coaxial probe method

Additionally, an automated electronic calibration refresh feature recalibrates the system automatically,
in seconds, before each measurement is made. This virtually eliminates cable instability and system
drift errors.

The coaxial probe method is convenient and operates over a wide 200 MHz to 50 GHz frequency range.
It is not well suited to low loss materials, magnetic materials, or where high accuracy is desired. Keysight
offers three probe designs. Each has unique strengths and limitations.

“High-Temperature Probe” has the lowest frequency coverage of all the probes. It can be used with an
impedance analyzer down to 10MHz. The large flange makes it easier to measure solid materials.

“Slim Form Probe” is a low-cost consumable design. It is the lowest cost of all the probes. It is ideal for
measuring materials that would destroy the probe, such as curing epoxy or cement. Because the tip is so
narrow and the tip is not as flat as the other probes, it is not suitable for measuring hard solids.

“Performance Probe” combines rugged high-temperature performance with high-frequency performance

all in one slim design. It withstands extreme temperatures like the high-temperature probe and covers up
to 50GHz like the slim form probe. In addition, it is hermetically sealed on both ends making it ideal for
applications that need sterile equipment. This probe can be autoclaved.

There are few assumptions while making this measurement. The technique assumes that the sample is
semi-infinite or endless, or at least as far as the network analyzer can see. How far into the material the
field extends depends on the material properties, the frequency of the measurement, and the dimensions
of the probe. Air gaps between a solid and the probe tip, and bubbles in a liquid will cause errors. The
sample is also assumed to be non-magnetic and isotropic.

Methods for Measuring the Dielectric Properties of Materials 152

 15.9 4.8

17.0

3.5 mm 16.8 3.0

connector
47.1 19.0
All Dimensions in mm

22 mm Finish, Nickel 100 U inches

200 mm
Borosilicate
150 Glass Seal
1.6 9.5

Nickel Plated
2.4 mm Male Stainless Steel Tungsten Center
connector T304L Body Conductor
Stainless Steel
T304L Tip
Figure 15.5. Coaxial probes

Transmission Line Method

A typical transmission line system consists of a vector network analyzer and a sample holder connected
between the two network analyzer ports. The software calculates permittivity from calibrated full two-
port S-parameter measurements. Keysight offers this software and several varieties of transmission line
sample holders.

The material sample is assumed to fill the cross-section of the fixture with no air gaps, have smooth
flat faces, and be uniform throughout. Coaxial airline fixtures are broadband, but the samples are more
difficult to machine. Waveguide fixtures extend to the mm-wave frequencies and the samples are simpler
to machine, but their frequency coverage is banded. Because the coaxial and waveguide transmission
line size scales with frequency, the practical sample size determines the frequency limits. Waveguides at
frequencies much lower than 5GHz start to get very large, and a large sample is needed. Below 1GHz for
many materials, sample length also becomes an issue. At frequencies above 75GHz, both coaxial and
waveguide dimensions get very small, and it becomes too difficult to machine the tiny samples.

Methods for Measuring the Dielectric Properties of Materials 153

 Coaxial

Waveguide
Reflection Transmission
(S11 ) (S 21 )

Heating panels Thermal

insulation
Furnace

Thermocouple
Sample

Figure 15.6. Transmission line method

Transmission line sample holders can be made from coaxial airlines or waveguide straight sections. Both
are widely available in different frequencies from Keysight and other connector manufacturers. Samples
must fit inside. This technique works best for hard solids that can be machined. It is possible, although
more difficult to contain liquids and powders inside these. Coaxial sample holders offer broadband
frequency coverage; however, it is more difficult to machine solid materials to the shape needed to fit
inside. Waveguide straight sections offer banded frequency coverage, but it is much easier to machine
solid materials to fit inside.

Table 15.1 shows the algorithms used in the N1500A material measurement suite from Keysight
technologies. They will convert the measured S-parameters to permittivity or permeability. The first
three require a two-port fixture. The last two require a one-port fixture which may be better for liquids or
powders where a shorted waveguide section can be turned on end and filled. One-port fixtures may also
be better for measurements at high temperatures where one end of the waveguide can be heated while
cooling mechanisms keep the network analyzer cool.

A special variant of the transmission line method is “free space” and is best for high-temperature
measurements since the sample is not enclosed in any kind of fixture, see Figure 15.6. The MUT is
assumed to be large, flat, and uniform throughout. The free space antennas are connected to a vector
network analyzer that measures the reflection and transmission from the MUT which are then converted
to permittivity and permeability. There are many free space measurement methods available to choose

Methods for Measuring the Dielectric Properties of Materials 154

from. Free space techniques use antennas to focus microwave energy through a slab of material without
the need for a test fixture. The same algorithms that are used for the transmission line technique can be
applied to free space.

Optimum
Algorithm Measured Output
Length

Nicolson-Ross S11,S21,S12,S22
λg /4 εr and µr
(PN 8510-3) or S11,S21)

Precision (NIST) S11,S21,S12,S22 nλg /4 εr

S21,S12
Fast nλg /4 εr
(S21)

Short-circuited
S11 λg /2 εr
back

Arbitrary dielectric
S11 λg /2 εr
back

Table 15.1. Algorithms supported in N1500A material measurement suite

Resonant Cavity Method

The resonant cavity method is best suited for thin films, substrate materials, and other low-loss dielectric
materials. The resonant cavity method uses a network analyzer to measure resonant frequency and Q
of a resonant cavity fixture, first empty and then loaded with the sample under test. Permittivity can then
be calculated from these measurements, knowing the volume of the sample and some other parameters
about the resonant cavity. As it is a resonant method, only one frequency point is reported.

However, it is much more sensitive and has a better resolution than the other techniques. The typical
resolution for this method is 10-4 whereas the broadband method is 10-2. A least-squares circle fitting
technique is used to calculate Q, which uses both magnitude and phase information and is more
repeatable than other Q calculation methods. The software then calculates εr’, εr”, and loss tangent and
displays them in its easy-to-use interface.

Methods for Measuring the Dielectric Properties of Materials 155

Figure 15.7. Resonant cavity method with network analyzer

Material characterization is a specialized and complex field of measurement. It has applications

in numerous industries. Choice of measurement techniques depends on several factors and it is
extremely important to choose the right method for your measurements to be able to achieve accurate
measurements. Keysight provides comprehensive material characterization solutions and can act as a
one-stop-shop of measurement gear, fixtures, and material measurement software.

Methods for Measuring the Dielectric Properties of Materials 156

Section 4

Radar and EW Subsystem Test

Introduction
For engineers and scientists, the names behind the earliest experiments in electromagnetism are
part of our everyday conversations: Heinrich Hertz, James Clerk Maxwell, and Nikola Tesla. Fast
forward from their work in the late 19th and early 20th centuries to the early 21st Century: the
fundamental concept—metallic objects reflect radio waves—has evolved into a host of technologies
that are pushed to the extremes in military applications: detecting, ranging, tracking, evading,
jamming, and more.

As is the case in commercial electronics and communications, the evolution from purely analog
designs to hybrid analog/digital designs continues to drive advances in radar system capability
and performance. Frequencies keep reaching higher and signals are becoming increasingly agile.
Signal formats and modulation schemes—pulsed and otherwise—continue to become more
complex, and this demands wider bandwidth. Advanced digital signal processing (DSP) techniques
are being used to disguise system operation and thereby avoid jamming. Architectures such as
active electronically steered arrays (AESA) rely on advanced materials such as gallium nitride
(GaN) to implement phased-array antennas that provide greater performance in beamforming and
beam steering.

Within the operating environment, the range of complexities may include ground clutter, sea clutter,
jamming, interference, wireless communication signals, and other forms of electromagnetic noise.
It may also include multiple targets, many of which utilize materials and technologies that present a
reduced radar cross-section.

This section moves ahead from component level to subsystem level characterization. Starting
with the basics of radar and EW transmitter and receiver, it covers new age requirements,
technologies and instruments to measure those. The section also highlights the methodology to
test Transmit-Receive (TR) modules efficiently and concludes with understanding different
techniques of antenna testing.
Chapter 16
Signal Source Essentials- Parameters, Technologies, and Methods
The essence of radar is the ability to gather information about a target — location, speed, direction,
shape, identity, or simply presence. This is done by processing reflected radio frequency (RF) or
microwave signals in the case of primary radars, or from a transmitted response in the case of
secondary radars.

In most implementations, a pulsed-RF or pulsed-microwave signal is generated by the radar

system, beamed toward the target in question, and collected by the same antenna that transmitted
the signal. This basic process is described by the radar range equation. The signal power at the
radar receiver is directly proportional to the transmitted power, the antenna gain (or aperture size),
and the radar cross-section (RCS) (i.e., the degree to which a target reflects the radar signal).
Perhaps more significantly, it is indirectly proportional to the fourth power of the distance to the
target. Given the large attenuation that occurs while the signal is traveling to and from the target,
having high power is very desirable; however, it is also difficult due to practical problems such as
heat, voltage breakdown, dynamic power requirements, system size, and, of course, cost.

Signal Source Essentials- Parameters, Technologies, and Methods 159

Radar Signal Parameters
Every radar transmission starts with a known carrier signal and today these typically operate at microwave
or millimeter-wave frequencies. The carrier may be coded, in part because changes in the return signal
provide a more accurate measurement of the distance to the target. In most cases, pulse modulation
is also applied because, controlling pulse duration, repetition rate and power, enhances the resolution
and maximum range of the radar system. Radar signals can be defined in the Time Domain, Frequency
Domain, and in terms of Pulse Descriptor Words (PDW) that are defined by amplitude, frequency, and
timing information. The advantage of PDWs is that they are an efficient way to store and digitally stream
radar signals.

Optimizing Pulse Parameters

The characteristics of a pulsed radar signal largely determine the performance and capability of the
system. Pulse parameters such as power, repetition rate, width, and modulation are traded off to obtain
the optimum combination for a given application (Figure 16.1). Pulse power directly affects the maximum
range of detection. Droop across the pulse top indicates instability in the output power.

Figure 16.1. The essential characteristics of real-world pulses affect the performance of the radar system.

The time between pulses—the pulse-repetition interval (PRI)—determines the maximum unambiguous
range to the target; pulse repetition frequency (PRF) is the inverse of PRI. The duration of the pulse—the
pulse width—determines the spatial resolution of the radar: pulses must be shorter than the time it takes
for the signal to travel between the target details; otherwise, the pulses overlap in the receiver.

Together, pulse width and pulse shape determine the spectrum of the radar signal. Decreasing the pulse
width increases signal bandwidth; however, wider system bandwidth results in higher receiver noise for
a given amount of power, and this reduces sensitivity. Also, the pulse spectrum may exceed regulated
frequency allotments if the pulse is too short.

Signal Source Essentials- Parameters, Technologies, and Methods 160

The shape can be the familiar trapezoidal pulse with rapid but controlled rise and fall times; it may also
be an alternative shape such as Gaussian and/or raised-cosine. Because the pulse shape can determine
the signal bandwidth and also affect the detection and identification of targets, it is chosen to suit the
application requirements.

Short pulses with a low repetition rate maximize resolution and unambiguous range, and high pulse
power maximizes the radar’s range in distance. There are, however, practical limitations in generating
short, high-power pulses. For example, higher peak power will shorten the life of tubes used in high-
power amplifier designs. This conundrum would be a barrier to increasing radar performance if radar
technology stopped here. However, complex waveforms and pulse-compression techniques can be used
to greatly mitigate the power limitation on pulse width.

Adding Pulse Compression

Compression techniques allow relatively long RF pulses to be used without sacrificing range or resolution.
The key to pulse compression is energy: using a longer pulse reduces the peak power of the transmitted
pulse but maintains the same average pulse energy. The received pulse is compressed using a match-
correlation filter, producing a shorter pulse of greater peak power and narrower width.

From this, a pulse-compression radar realizes many of the benefits of a short pulse: improved resolution
and accuracy; reduced clutter; better target classification; and greater tolerance to some electronic
warfare (EW) and jamming techniques. One area that does not improve is minimum range performance:
the long transmitter pulse may obscure targets that are too close to the radar.

The ability to compress the pulse with a matched filter is achieved by modulating the RF pulse in ways
that facilitate the compression process. The matching filter function can be achieved digitally using the
cross-correlation function to compare the received and transmitted pulses. The sampled receive signal is
repeatedly time-shifted, fast Fourier transformed and multiplied by the conjugate of the Fourier transform
of the sampled transmit signal (or a replica).

The output of the cross-correlation function is proportional to the time-shifted match of the two signals.
A spike in the cross-correlation function or matching-filter output occurs when the two signals are
aligned. This spike is the radar return signal, and it may be 1000 times shorter in duration than the
transmitted pulse.

Even if two or more of the transmitted pulses overlap in the receiver, the sharp rise in output occurs only
when each pulse is aligned with the transmit pulse. This restores the separation between the received
pulses and, with it, the range resolution. To reduce the time-domain sidelobes created during the cross-
correlation process, the received waveform can be processed with a windowing function of Hamming
shape or similar.

Signal Source Essentials- Parameters, Technologies, and Methods 161

Ideally, the correlation between the received and transmitted signals would be high only when the
transmitted and received signals are exactly aligned. Many modulation techniques can be used to
achieve this goal: linear FM sweep, binary phase coding (e.g., Barker codes), or polyphase codes (e.g.,
Costas codes).

Accounting for Doppler Effects

Most targets of interest are moving. This causes the frequency of the returned signal to be shifted higher
if the target is moving toward the radar and lower if the target is moving away. Unfortunately, this Doppler
frequency shift can reduce the sensitivity of location detection.

As mentioned above, the output of the cross-correlation filter is proportional to the match between the
received and transmitted signals. If the received signal is slightly lower or slightly higher in frequency, then
the filter output is somewhat lower.

For a simple pulse, the filter response follows the familiar sin(x)/x shape as a function of Doppler
frequency. In extreme cases, the frequency of the received signal may shift far enough to correlate with a
sidelobe of the transmit signal.

Note that short pulses have a relatively wide initial lobe in the sin(x)/x response and so tend to be
“Doppler tolerant” compared to longer pulses. In pulse compression schemes such as Barker coding,
the matching-filter output drops off much faster than the sin(x)/x of the simple pulse, making them
“Doppler intolerant.”

Doppler shifts in linear FM pulses can create an error in the location information because the highest
cross-correlation occurs where the swept frequencies in the received pulse are best aligned with the
swept frequencies in the transmit pulse. This offset is directly proportional to the Doppler shift.

Signal Source Essentials- Parameters, Technologies, and Methods 162

Bringing it all Together
Graphs called ambiguity diagrams illustrate the performance of different pulse-compression schemes
as a function of pulse width and Doppler frequency shift (Figure 16.2). Even though Doppler shift can
reduce detector sensitivity and cause errors in time alignment, it also provides important information
about the target.

Pulse type Frequency domain Ambiguity diagram

0 dB Doppler
frequency

Short RF pulse Pulse

width*

16 MHz

–3 dB Doppler
frequency

Long RF pulse Pulse

width*

8 MHz

–21dB Doppler
frequency
Pulse
Swept RF pulse width*

1.2 GHz

–20 dB Doppler
frequency
13-bit Barker Pulse
coded RP pulse width*

1.2 GHz

*Figure
The term
16.2. pulse width on
The ambiguity the ambiguity
diagrams illustratediagram refers toversus
location accuracy the pulse width
Doppler at the This
accuracy. radarfigure
detector
showsoutput.
relative
ambiguity diagrams for different types of radar pulses.

*Note: in the ambiguity diagram, “pulse width” refers to the width at the output of the radar detector.

Signal Source Essentials- Parameters, Technologies, and Methods 163

Signal Generator Essentials
Engineers use signal generators to test components, receivers, and systems for a variety of applications
throughout the product development cycle. The output signal can be as simple as a continuous wave
(CW) or complex, like a digitally modulated signal. Figures 16.3 and 16.4 show common signal generator
use cases for component and receiver tests.

Component RF transceiver (Rx)

x System Component
(stimulus-response)
RF Signal
I+, I-
RF / IF RF / IF
RF Rx
Q+, Q- RF / IF I Q
DUT
Baseband
BBIQ

Intermodulation (2-tone) RF transceiver (T

( x)
x Component substitution
CW
CW #1
I+, I-
CW #2
∑ RF Tx RF / IF
RF / IF

X
DUT
Upconvert
r er / downconvert
r er

Figure 16.3. Signal generator use cases for component characteristic tests or a system component

Sensitivity CO-channel rejection CO-channel selectivity

In-band CW / modulated Adjacent / alternate channel
interfering interfering
Desired
signal Desired Desired
BERT
Rx signal ∑ signal ∑
DUT BERT BERT
Data / clock
Data / clock Rx Data / clock Rx
DUT DUT

Spurious immunity Fading measurement

Out-of-band CW / modulated Out-of-band CW / modulated
Desired
interfering interfering signal

Desired Desired Channel

signal ∑ signal ∑ Simulator
BERT BERT
Data / clock Rx Data / clock Rx
DUT DUT Rx
DUT

Figure 16.4. Signal generator use case for receiver sensitivity tests

Signal Source Essentials- Parameters, Technologies, and Methods 164

Signal generators can be classified based on their form factor and capabilities.

Form Factor: Benchtop or Modular?

The most common signal generator form factor is the benchtop. We typically see these boxed
instruments on benches and in racks. Benchtop signal generators are well-suited for R&D, where
engineers use the front panel controls to analyze and troubleshoot devices.

The PXIe modular form factor signal generators are compact instruments housed in a PXIe chassis and
controlled using a PC. Several PXIe signal generators can be placed in a single chassis, making them
ideal for applications that require multi-channel measurement capabilities, fast measurement speed,
and a small footprint. A PXIe signal generator often uses the same software applications as a benchtop
signal generator, providing measurement consistency and compatibility from product development to
manufacturing and support.

Figure 16.5. Benchtop and PXI modular signal generator

Capabilities: Analog, Vector, and Agile Signal Generators

Analog signal generators supply sinusoidal continuous wave (CW) signals with optional capability to add
AM, FM, ΦM, and pulse modulation. The maximum frequency range for analog signal generators spans
from RF to microwave. Most generators feature step/list sweep modes for passive device characterization
or calibration.

Vector signal generators (VSG), a more capable class of signal generators, enable complex digital
modulation schemes. VSGs have a built-in quadrature (also called IQ) modulator to generate complex
modulation formats such as quadrature phase-shift keying (QPSK) and 1024 quadrature amplitude
modulation (QAM). When combined with an IQ baseband generator, virtually any signal can be emulated
and transmitted within the information bandwidth supported by the system.

Optimized for speed, agile signal generators can quickly change frequency, amplitude, and phase of the
signal. They also have the unique capability to be phase coherent at all frequencies, at all times. This
attribute, along with extensive pulse modulation and wideband chirp capabilities, make them ideal for
electronic warfare and radar applications.

Signal Source Essentials- Parameters, Technologies, and Methods 165

Overview of Key Specifications
To select the right signal generator for your project, you’ll need to understand its performance
specifications. Specifications tell you about the capability of your signal generator. Let’s explore major
specifications: frequency, amplitude, and spectral purity performance.

Frequency specifications
The frequency specification defines the range, resolution, accuracy, and switching speed of your
signal generator.

• Range — the maximum and minimum output frequencies your signal generator can output.
• Resolution — the smallest frequency change.
• Accuracy — how close the source’s output frequency is to the set frequency.
• Switching — how fast the output settles down to the desired frequency.

Figure 16.6. Spectrum analysis with frequency and amplitude readouts

Signal Source Essentials- Parameters, Technologies, and Methods 166

Amplitude specifications
Amplitude specifications include range, resolution, and switching speed.

• Range — the difference between the maximum and minimum output power capability of the
signal generator. The signal generator’s output attenuator design determines its range.
The output attenuator allows the signal generator to produce extremely small signals used to test
a receiver’s sensitivity.
• Resolution — the smallest possible power increment.
• Switching speed — how fast the source can change from one power level to the next.

Pmax Accuracy
Power

Dynamic Output power

capability of the
range signal generator

Pmin

Frequency

Figure 16.7. Power output range and accuracy

Spectral purity
Spectral purity is the inherent stability of a signal. A perfect signal generator will generate a sinusoidal
wave at a single frequency without the presence of noise. However, signal generators consist of non-ideal
components which introduce noise and distortion. The specifications associated with spectral purity are
often the most difficult to understand. These specifications include phase noise, harmonics, and spurs as
shown in Figure 16.8.
CW
output

Harmonic spur
Phase
~30dBc from
noise Non-harmonic spur
non-linear
Sub-harmonics (dBc/Hz) from power supplies
components
from multipliers used to from LO’s and other contributors
extend the frequency
output

Broadband noise floor

Thermal noise of source
0.5 f0 f0 2 f0
Figure 16.8. Various non-ideal spectral components

Signal Source Essentials- Parameters, Technologies, and Methods 167

 • Phase noise — a frequency-domain view of the noise spectrum around the oscillator signal. It
describes the frequency stability of an oscillator.

• Harmonics — integer multiples of the sinusoidal fundamental frequency output. These harmonics are
caused by the non-linear characteristics of components used in the signal generator.
• Spurs — non-random or deterministic signals created from mixing and dividing signals to get the
carrier frequency. These signals may be harmonically or non harmonically related to the carrier.

Ultra-Wideband Arbitrary Waveform Generators (AWGs)

Arbitrary waveform generators can vary in regards to sample rate and resolution, and those two
parameters are often inversely related (Figure 16.9). AWGs with higher sample rates will have lower
resolutions than those with lower sample rates and vice-versa. AWGs can be used to simulate high-
density radar signals and communications signals within their bandwidths.

PCI-Express or USB 2.0 link from

external PC or embedded controller

SYNC MRK OUT

Sync Clk SAMPLE MRK OUT
Sample Memory
(up to 2 GSa)

Channel DIRECT OUT

Module DAC
FPGA
FPGA
Sequence Memory
(512 K entries)
Sync Clk Sample Clk
Triggen Event In

Interface transferring
Var. delay

Var. delay

Dynamic Control In
48.64 DAC samples or
24 IQ sample pairs AC Amp.
per Sync Clk
AMP OUT
Ref CLK In
Sample CLK In Clock
Generation DC Amp.

InL Clk Channel 1

Channel 2

SYNC CLK OUT (=sample clock/48 or sample clock/64 or

sample clock/(24’interpolation factor)

Figure 16.9. Modern Arbitrary Waveform Generators are comprised of much more than DACs. Such new
capabilities involve memory sequencing, clock sharing for synchronization, and different output paths to optimize
signals depending on the application. Shown here is the block diagram for the high-resolution M8190A Arbitrary
Waveform Generator.

Signal Source Essentials- Parameters, Technologies, and Methods 168

Ultra-wideband AWGs have extremely wide bandwidth, which allows for the generation of single and
multiple emitters across wide frequency spans. High-resolution AWGs can output signals with a high
dynamic range within narrower bandwidths. Key capabilities include support for simultaneous pulses
(pulse-on-pulse) and the ability to modify I/Q data to allow for environmental effects. Multiple channels
can also be synchronized rather easily for tests requiring multiple coherent channels. In contrast to other
source types, ultra-wideband AWGs tend to have lower resolution and dynamic range. They also require
a large amount of storage due to their extremely high sample rates. High-resolution AWGs have lower
bandwidths and can be upconverted using the appropriate architecture.

a) Baseband Generation

2-Channel AWG Quadrature Modulator

I Q

IF/RF Out

Lowpass Filter

b) Direct RF Generation in the First Nyquist Band

IF/RF Out
1-Channel AWG

Figure 16.10. AWGs can generate radar (and any RF) signals following to different basic schemes

Electronic Warfare Signal Generation Technologies and Methods

Productive and efficient engineering of electronic warfare (EW) systems requires the generation of
test signals that accurately and repeatedly represent the EW environment. Simulation of multi-emitter
environments is vital to ensure realistic testing.

Simulation for these multi-emitter environments traditionally encompasses large, complex, and custom
systems during the system qualification and verification stage. These systems are usually not widely
available to EW design engineers as R&D test equipment. EW designers working on optimization and
pre-qualification are at a disadvantage in comparison to wireless engineers performing similar tasks. EW
engineers often discover the nature and magnitude of performance problems later in the design phase —
leading to delays, design rework, and solutions that are not optimal.

Signal Source Essentials- Parameters, Technologies, and Methods 169

Realism and fidelity in multi-emitter environments
Validation and verification of EW systems are heavily dependent on testing with realistic signal
environments. Adding high-fidelity emitters for greater signal density creates a realistic EW test
environment. In addition, emitter fidelity and density, platform motion, emitter scan patterns, receiver
antenna models, the direction of arrival, and multipath and atmospheric models enhance the ability to
test EW systems under realistic conditions. The designs for modern EW systems can identify emitters
using precise direction finding and pulse parameterization in dense environments of millions of pulses
per second.

The cost of the test is as important as test realism, as the relationship between cost and test fidelity
is exponential. As test equipment becomes more cost-effective and capable, more EW testing can be
performed on the ground — in a lab or chamber — rather than in flight. Even though flight testing can
add test capability, it does so at a high cost. It is typically done later in the program lifecycle, adding risk
and further expense to the program through missed deadlines if the system under test (SUT) fails. It is far
better to test early in a lab environment with as much realism as possible, where tests are easily repeated
to identify iteratively and to resolve issues.

Challenges of simulating multi-emitter environments

The modern spectral environment contains thousands of emitters — radios, wireless devices, and tens
to hundreds of radar threats — producing millions of radar pulses per second amidst background signals
and noise. Figure 16.11 shows a general overview of the threat frequency spectrum.
Pulse density (log)

Acquisition, GCI

Fire control
Early warning

VHF UHF L S C X Ku K Ka

A B C D E F G H I J K

Figure 16.11. A general representation of the threat density vs. frequency band in a typical operational environment.
The full RF/microwave environment would be a combination of the threat and commercial wireless environments.

Signal Source Essentials- Parameters, Technologies, and Methods 170

Simulating this environment is a significant challenge — especially in the design phase when design
flexibility and productivity are at their greatest. The situation is quite different from the typical wireless
design task, where a single signal generator can produce the required signal, augmented by a second
signal generator to add interference or noise.

In EW design, the multiplicity and density of the environment — and often the bandwidth — make it
impractical to use a single source or a small number of sources to simulate a single emitter or a small
number of emitters. Cost, space, and complexity considerations rule out these approaches.

The only practical solution is to simulate many emitters with a single source, and to employ multiple
sources — each typically simulating many emitters — when required to produce the needed signal
density or to simulate specific phenomena such as angle-of-arrival (AoA).

The ability to simulate multiple emitters at multiple frequencies depends on the following: pulse repetition
frequency; duty cycle; number of emitters; and the capability of the source to switch between frequency,
amplitude, and modulation quickly.

A source’s agility is a factor in its ability to simulate multiple emitters. Source frequency, phase, and
amplitude settling time (whichever is greater) is the transition time between playing one pulse descriptor
word (PDW) and the next.

Improvements simplify integration and reduce cost

Simulating more threats to create higher pulse density requires more parallel simulation channels — even
if the simulation channel can switch frequency, phase, and amplitude quickly. This is because pulses
begin to collide in the time domain as the number of emitters, their PRFs, and their duty cycles grow
larger1. Pulse that overlap in the time domain must be played out of parallel generators or selectively
dropped based on a PDW priority scheme. Unfortunately, the increased realism of a higher-density
environment comes at a substantially higher system cost, as shown in Figure 16.12.
Cost

Value

Legacy simulation
technology Modern
simulation technology

Fidelity
Figure 16.12. Simulation fidelity and cost increase exponentially. System integrators and evaluators must determine
the level of cost versus fidelity to ensure system performance. New simulation technologies enable more simulation
realism and fidelity at a lower cost.

Signal Source Essentials- Parameters, Technologies, and Methods 171

In the past, simulations have generally been created with a separate component for each emulation
function, such as signal generation, modulation/pulsing, attenuation or amplification, and phase shift. The
same PDW would be sent to each functional component to provide output on a pulse-to-pulse basis. For
instance, a synthesizer would generate the output frequency, while a separate modulator would create
pulsed modulation or AM/FM/PM modulation. Amplifiers and attenuators would adjust the signal to the
desired output power level. Figure 16.13 is an example of the system’s topology.

PDWs Control parameters

Pulse RF Amplifier/ EW simulation

generator generator attenuator output

Synchronization
Figure 16.13. In the traditional approach, PDW control parameters are sent in parallel to multiple functional
elements, on a pulse-to-pulse basis, to generate and modify the desired signal. This approach results in a complex
system, demanding precise synchronization.

Because multiple functional components are required to produce each output channel, time
synchronization is a significant configuration and operational challenge. A wide variety of settling times
and latencies must be fully characterized to optimize pulse density by minimizing lockout periods.

This approach can be scaled directly to create multiple coordinated channels, as shown in Figure 16.14
However, systems configured in this way require a large footprint — occupying more rack space — and
cost escalates quickly.

Modulation RF Signal
Generator Generator Attenuator

Modulation RF Signal
Generator Generator Attenuator
Distribution
Network

Emulated
RF Output
Control Modulation RF Signal to EW
Generator Generator Attenuator Receiver

Modulation RF Signal
Generator Generator Attenuator

Figure 16.14. A signal generation approach using separate functional elements can be scaled up to increase pulse
density and generate a more realistic environment. The cost and space requirements scale up rapidly as well2.

Signal Source Essentials- Parameters, Technologies, and Methods 172

The controller in Figure 16.14 would route PDWs to channels based on emitter parameters, such as
frequency, amplitude, and pulse repetition frequency, and the availability of each channel to implement the
PDW. Because a channel cannot execute the parameters of two different PDWs at the same time, one
could be shunted to a backup channel or dropped according to its priority.

EW receivers must be able to handle millions of pulses per second, where most of the pulse density
occurs at the X-band. EW receivers must be able to handle pulses arriving at the same time at different
frequencies from different angles. Creating pulses that are coincident with one another in the time domain
should be a goal of simulation to increase simulation realism.

Though Figure 16.14 describes a very capable system, the system elements are not highly integrated.
Recent developments in analog and digital signal generation technologies are enabling a higher degree of
integration and solutions which are more cost- and space- efficient.

There are several methods of controlling simulations, depending on test objectives. Figure 16.14 shows
systems with a traditional, distributed architecture. The synchronization of an agile local oscillator
(LO) with functions such as pulse modulation, frequency/phase modulation, and amplitude control is
a considerable challenge. In an integrated EW test solution such as the UXG, this synchronization is
automatic, provided by the test equipment itself. By simplifying hardware and system complexity, this
integrated approach promises to improve both performance and reliability.

Control of hardware-in-the-loop testing

Depending on the integration of simulation elements and the simulation length, scenarios can be played
from list memory or streamed over a digital interface such as LAN or low-voltage differential signaling
(LVDS). List mode plays PDWs from list memory for shorter scenario lengths with some ability to trigger
between lists for an adaptive (closed-loop) simulation in response to the SUT.

For example, there is often a need to switch between one simulated threat mode to another in response
to identification and jamming by the SUT. For long scenario lengths with fast control over scenario
changes, PDWs can be streamed over the LAN to the signal generation system operating in an agile
controller mode. In this case, simulation software generates batches of PDWs according to simulation
kinematic granularity and streams them ahead of their desired playtime.

The goals are to stress the SUT with increasing pulse density, depending on the number of simulation
channels available and the parameters of the threats to be simulated. As pulse density increases, PDWs
can be dropped according to a priority scheme as they increasingly collide in the time domain, and there
are insufficient signal generation channels to play them.

Creating AoA
In addition to creating emitters with the desired fidelity and density, it is also important to match the
geometry and kinematics of EW scenarios. This is because the AoA of a radar threat to the EW system
changes slowly compared to other parameters, such as center frequency and pulse repetition frequency.

Signal Source Essentials- Parameters, Technologies, and Methods 173

EW systems measure AoA and estimate distance using amplitude comparison, differential Doppler,
interferometry (phase difference), and time difference of arrival (TDoA). Precise AoA measurements enable
precise localization of radar threats. New stand-off jamming systems use active electronically scanned
arrays capable of precise beamforming to minimize loss of jamming power due to beam spreading toward
a threat. EW receivers with better AoA capability reduce the need for pulse deinterleaving and sorting.
Consequently, AoA is an increasingly important test requirement.

Techniques for creating AoA

In the past, AoA was created with a combination of signal sources and analog phase shifters, attenuators,
and gain blocks in the cable path to the SUT. Analog elements in the cable path took up space, had
limited resolution, and were expensive.

As an alternative, and depending on their architecture, sources can be linked together to create
phase-coherent output, allowing for exceptional control over creating phase fronts to the SUT.
Similarly, amplitude control at the source can be used to create appropriate amplitude differences at
SUT receive channels.

The ability to control AoA to meet modern test requirements depends on the architecture of the source.
At a minimum, it should be possible to lock the LOs of multiple sources together so that they all share the
same phase. Often, calibration is required to align the phase and timing between sources.

Creating small, accurate, and repeatable differences in phase or frequency between channels is the next
challenge. Sources based on a direct digital synthesis (DDS) architecture allow AoA to be controlled
digitally in a numerically controlled oscillator. Phase alignment in a DDS source is then a matter of sharing
reference clocks. Calibrations to provide accuracy and repeatability can be uploaded to a table to be
applied in real-time.

Overview of source technologies for EW test

The characteristics and tradeoffs of EW signal generation systems are primarily determined by the core
synthesizer and oscillator technologies used. This section summarizes the key technologies currently
available:

• Direct analog synthesis (DAS)

• A phase-locked loop or indirect analog synthesis (PLL, frequently fractional-N) – Direct
digital synthesis

General source requirements

Signal sources used to test EW systems must be broadband. Traditionally, a frequency range of 0.5 to 18
GHz was required. Frequency requirements have expanded dramatically in recent years, now beginning
near DC and extending as high as 40 GHz. They allow systems to simulate an early warning, fire control,
and missile-seeking radars from a single output channel.

Signal Source Essentials- Parameters, Technologies, and Methods 174

In addition to wide frequency coverage, sources for the EW test must have fast frequency, phase,
and amplitude switching speeds to simulate different radars operating in different modes in various
frequency bands.

PLLs and fractional-N synthesis

Indirect synthesis
Most general-purpose sources today are PLL-based, where a broadband oscillator such as a
voltage-controlled or YIG-tuned oscillator is locked to a stable reference in a phase-locked loop (PLL).
The PLL improves signal quality by reducing phase noise and spurious signals in the output. PLL-based
sources have been configured with a combination of sum and step loops or a single-loop with a fine
fractional division capability. These fractional-N PLLs offer excellent signal quality and fine frequency
resolution in a cost-effective single-loop configuration, making them an excellent choice for general-
purpose signal sources.

The required control loop filtering in PLLs results in a significant settling or loop response time. This
looping limits the ability of the synthesizer to switch frequency quickly. Due to their comparatively high
transition time, these sources are limited in their ability to simulate multiple radar threats out of a single
channel, even if they have the necessary broadband frequency coverage and frequency resolution. They
also lack phase-repeatable switching capability.

Direct analog synthesis

A direct analog synthesizer typically contains several stable frequency references multiplied or divided
from the same crystal oscillator reference. These frequency references (and their harmonics) can be
switched in and out of the signal path and multiplied, divided, added, and subtracted to provide fine
frequency resolution quickly. The frequencies of these references are chosen to reduce the number of
multiplication stages required, such that phase noise increases only moderately as the frequency is
increased. The division to lower frequencies reduces the phase noise.

Since the switches and arithmetic operators used in the DAS approach operate very quickly and do not
need loop filtering, these synthesizers have very high-frequency agility. They are a typical architecture for
traditional EW test solutions.

However, DAS technology has several drawbacks. First, numerous stages are required to achieve the
desired frequency resolution. Switching parallel and series multiplication, division, and mixing stages
require more hardware than PLLs and reduces reliability. Second, circuit noise from each stage is
cascaded, and phase noise is multiplied through the stages. Finally, each stage adds components that
increase size, weight, and cost.

On the positive side for EW applications, DAS has the potential for limited phase-repeatable frequency
switching. All frequencies are usually derived from the same reference, but divider ambiguities generally
preclude full phase-coherent switching.

Signal Source Essentials- Parameters, Technologies, and Methods 175

DDS now suitable for EW applications
The DDS approach, based on DAC circuits, is a natural fit for the needs of EW signal simulation. However,
until recently, DACs were not available with the required combination of fast sample rates and high purity.

Fast sample rates are needed to produce outputs with very wide bandwidth so that a minimum of
multiplying stages can be used to create the desired output frequencies. The use of either many
multiplying stages or a DAC of insufficient purity would limit the effective spurious-free dynamic range
(SFDR) of the EW synthesizer.

In concept, a DDS is one of the simplest types of signal generators. In a frequency-tunable DDS, data
from a numerically controlled oscillator is converted to analog form by a DAC and low-pass filtered to
remove image frequencies and harmonics. A block diagram of the key elements of a DDS is shown in
Figure 16.15.

Frequency Numerically Reconstruction

Analog
control controlled DAC lowpass
output
register oscillator filter

Reference oscillator Fclk

Figure 16.15. Principal functional blocks of a direct digital synthesizer

Advantages of DDS
The Keysight UXG agile signal generator uses DDS technology made possible by a proprietary DAC to
generate multi-emitter simulations. DDS has several advantages over other synthesis technologies
for EW applications: Digital control of extremely fine frequency and phase tuning increments within a
single clock cycle.

In the UXG agile signal generator, the frequency resolution is one millihertz and phase resolution is
sub-degree. Fractional-N techniques can provide microhertz resolution, but frequency changes are
much slower due to PLL filtering. DAS techniques provide rapid frequency switching, but at a cost in
frequency resolution.

DAS techniques offer hop speed and frequency/phase repeatability only under limited conditions.
Modulation is created in the digital domain, providing numerical precision and repeatability.

There are other advantages to using DDS that are of interest to the EW engineer. Many DDSs employ a
digital modulator for amplitude, frequency, and phase modulation for the creation of digitally modulated
signals in the numerically controlled oscillator. Linear frequency modulated (LFM) chirps and Barker
codes can also be directly synthesized using the numerically controlled oscillator. Chirp bandwidth
depends on the bandwidth of the bandpass filters after each multiplication stage and whether the signal
is crossing a band.

Signal Source Essentials- Parameters, Technologies, and Methods 176

Advanced Threat Simulation
The DDS architecture provides many advantages for EW threat simulation. As radar threats grow more
advanced and sophisticated, threat simulation systems must produce high-fidelity reproductions of these
signals. These reproductions include shaped pulses with varying rise and fall times, non-linear chirps, or
custom modulation. Implementing these effects with traditional analog building blocks, such as pulse and
I/Q modulators, can prove challenging. Fortunately, modern digital I/Q baseband systems can accurately
generate these complex waveforms while minimizing distortion and spurious signals.

You cannot create some threat scenarios, such as AoA, with a single-channel source. Those scenarios
depend on properly synchronizing the outputs of two or more sources. By precisely controlling the
amplitude, phase, and time delay of each source output, you can simulate the direction of a radar wave-
front as it reaches the multiple antennas of an EW SUT. Accomplishing this feat with multiple signal
generators scales up the cost, often resulting in redundant hardware — adding size, weight, power
consumption, and complexity.

The Keysight UXG agile vector adapter works in conjunction with the UXG agile signal generator. Figure
16.16 shows the block diagram.

UXG agile signal generator

6 GHz reference
8 - 18 GHz LO

1.8 GHz
1.6 GHz BW

.05 - 40 GHz

PDWs
Proprietary DAC

UXG agile vector adapter

Figure 16.16. High-level block diagram of the UXG agile vector adapter

Signal Source Essentials- Parameters, Technologies, and Methods 177

The vector adapter utilizes the 6 GHz reference and agile LO signals from the DDS source to avoid
duplication of this hardware while adding a digital I/Q baseband system, upconverter, and electronic
attenuator. The baseband generator memory stores the complex pulse waveforms represented by I/Q
data points. This digital information feeds the DDS engine, where it is converted to an IF signal and
upconverted to an RF frequency. An electronic attenuator provides agile amplitude scaling of the signal.

Minimizing Distortion
Traditional analog I/Q baseband systems must be carefully tuned to minimize signal distortion caused by
phenomena such as IQ gain imbalance (where the gain in the I and Q channels is slightly different) and IQ
skew (where the I and Q paths are not precisely in quadrature). Figure 16.17 (left image) shows how these
imperfections create in-band distortion. these distortion products occur within the signal bandwidth,
hence cannot be filtered out.

A digital baseband architecture mathematically shifts the I and Q channels by 90 degrees and digitally
sums them before conversion to an analog IF signal. This technique greatly reduces the amount of in-
band distortion. Figure 16.17 (right image) shows how this technique provides a higher fidelity signal.

AWG or Signal Gen. Signal Generator. AWG Mixer/

I Multiplier
Memory D/A X Memory X /LO
~ ~
90° + 90° + D/A

Memory D/A
Q X Memory X IF
Digital signal Digital signal
Analog signal Analog signal

Figure 16.17. A comparison of analog and digital baseband architectures

Signal Source Essentials- Parameters, Technologies, and Methods 178

Agile amplitude signal control is crucial for realistic threat simulation, where multiple emitters may be at
different distances and transmit at different power levels. Mechanical attenuators cannot switch quickly
enough to keep up with scenarios that potentially generate millions of pulses per second. To provide agile
amplitude control, you can use the baseband generator’s digital-to-analog converter (DAC) to scale the
signal quickly. Depending on the vertical bits of resolution available in the DAC, this method can provide
40 to 55 dB of agile amplitude range.

This level of performance may impose severe limits on the threat scenario, which could require a higher,
agile dynamic range. Moreover, the DAC technique for amplitude control cannot attenuate any spurious
signals created further down the signal chain. The EW receiver under test must perform additional
signal processing to determine if the detected signal is a genuine threat signal or a spurious one that
can be ignored.

Configuring the Threat Simulation System

Depending on the SUT characteristics and the complexity of the desired threat scenario, you may need to
configure the threat simulation system to support the differing channel and port counts. A channel refers
to the ability to independently tune the threat signal to the desired frequency, while a port refers to the
actual output port of the RF signal.

If the scenario requires large numbers of pulses at multiple frequencies with minimal dropped pulses, you
might select the 4-channel, 1-port configurations in Figure 16.18. The outputs of each analog or vector
source are combined to a single RF port.

Analog Vector

4 Channel x 1 Port System

• 4 Pulses at Same Time
• 1 RF Connection

Figure 16.18. 4-channel, 1-port configurations

If the threat scenario calls for AoA measurements, you need a different configuration. Figure 16.19 shows
two different 1-channel, 4-port test configurations. All four sources are tuned to the same frequency.
But the amplitude, phase, and time delay of each RF output are individually controlled to simulate the
direction of the threat.

Signal Source Essentials- Parameters, Technologies, and Methods 179

 Analog Vector

1 Channel x 4 Port AOA System

• 1 Pulse at Same Time
• 4 RF Connections

Figure 16.19. 1-channel, 4-port configurations

More complex scenarios may demand more extensive configurations, such as the 3-channel, 4-port
setup in Figure 16.20. The three channels of this configuration provide high pulse density with pulse-on-
pulse capability, as well as multiple ports for AoA testing.

SUT

Figure 16.20. Configuration for 3-channel, 4-port

Signal Source Essentials- Parameters, Technologies, and Methods 180

A variety of traditional technologies have been used to generate the signals needed for effective EW
simulation. Each of these technologies has brought a different combination of benefits and challenges.
The highest-fidelity solutions have provided realistic simulations of the EW environment, but their
complexity and expense have limited their use.

Recent innovations in core hardware such as DACs and FPGAs have enabled new solutions with the
hardware simplicity and reliability of traditional test equipment. These solutions will provide dramatic
improvements in solution cost and size, bringing high-fidelity EW environment simulation to a much
earlier phase in the design process. Using realistic EW environment simulation at the optimization and
pre-verification stages of design will improve performance, speed the design process, and reduce
overall costs.

References
1. Philip Kazserman, “Frequency of pulse coincidence given in n radars of different pulse widths and
PRFs,” IEEE Trans. Aerospace and Electronic Systems, Vol. AES-6, p. 657-662, September 1970.
2. Reproduced by permission from David Adamy, EW 101: A First Course in Electronic Warfare,
Norwood, MA: Artech House, Inc., 2001. © 2001 by Artech House, Inc.

SIMULATE MULTI-THREAT SCENARIOS | N5193A UXG

INTRODUCING THE NEW VXG MICROWAVE SIGNAL GENERATORS AGILE SIGNAL GENERATOR

RADAR AND EW COHERENT SIGNAL GENERATION AND ANALYSIS -

AMPLITUDE WALKS

https://www.keysight.com/see/radar-ew-subsystem-test

Signal Source Essentials- Parameters, Technologies, and Methods 181

Chapter 17
Measuring Radar and EW Signals
A radar transmitter is the most costly component of the system with the highest power
consumption, most stringent cooling requirements, and greatest influence on system performance.

Measuring Radar and EW Signals 182

There are many different terms used when talking about power, as shown in Figure 17.1. Average power
is the power that is integrated over the complete time waveform (on time and off time) of the radar. If the
pulse width and PRF are not constant, the integration time must be long enough to represent all possible
variations in pulse parameters. Most typical RF and microwave power meters are average power meters
and respond to the heating energy of the signal. Peak power is the maximum instantaneous power. Pulse
power is the integrated or average power for one complete pulse.

Peak Pulse top

power amplitude Overshoot

Average power
Pulse
width
Pulse
Duty cycle delay

PRI PRF/PRR

Figure 17.1. Pulse parameters

Other parameters, including duty cycle, pulse width, PRF, and rise and fall times as shown in Figure 17.1,
are useful for characterizing the power of the radar signal.

From a radar equation standpoint, the power term corresponds to the power of the transmit pulse.
If the integration term is excluded, the equation applies to a single pulse. Therefore, it can be useful
to examine the peak and pulse power on an individual pulse basis. This technique is becoming more
important for modern radar systems in which pulse width and PRF are dynamically adjusted and the
pulse profile may be contoured to improve system performance. It is also becoming easier to perform
with modern test equipment.

It should be noted that average power measurements are a common method for characterizing the
power of a radar signal. These are simple to perform and require only low-cost instruments. If pulse
characteristics, such as the duty cycle of the radar signal, are known, the pulse power can be derived or
estimated based on average power. Note, however, that this derived result does not provide information
about droop, or any peak excursions that may occur due to ringing or overshoot. The result would be
nearly equivalent to the pulse-top amplitude, and in the case of a perfectly square pulse, it would be
equivalent to the true peak power or pulse power.

Along with measuring power, the spectrum shape is critical to verifying that a radar system is operating
efficiently. For example, an unsymmetrical or incorrect spectral shape indicates a radar that is operating
less than optimally. In such cases, the radar may be wasting power by transmitting or splattering power at
unwanted frequencies, causing out-of-band interference. For some radar systems, pulse shaping is used
to reduce the level of the spectral sidelobes, to improve the efficiency and life of radar components, and
to reduce bandwidth.
Measuring Radar and EW Signals 183
There are several options for measuring radar power, pulse characteristics, and spectrum including the
use of a power meter, signal/spectrum analyzer, or vector signal analyzer. Because each instrument
has advantages and limitations, the best choice is determined by the measurement objectives and the
constraints on the radar and the test instrument. This section will describe how to make measurements
with each of these instruments.

Maximum Instrument Input Level

One of the first things to consider is the magnitude of RF power that could be encountered. Parameters
such as frequency, antenna match (SWR), pulse width (PW), pulse repetition time (PRT), and duty cycle
will affect power measurements and selection of measurement hardware.

RF and microwave instruments are limited in both the amount of average power and the amount of peak
power that can be input without damaging the instrument. For typical radar systems, with pulse powers
of approximately 1 MW, a directional coupler is required to sample the transmitter power and provide a
safe drive level to the test instrument.

Measuring Pulse Power with a Power Meter

The most common (and lowest-cost) way to measure pulse power is with a power meter. The right
power meter can provide several measurements including average power, peak power, duty cycle, and
even power statistics. One of the first things to consider when measuring with a power meter is the
power sensor.

Figure 17.2. P-Series power meters

The power sensor

The power sensor converts high-frequency power to a DC or low-frequency signal that the power meter
can then measure and relate to a certain RF or microwave power level. The three main types of sensors
are thermistors, thermocouples, and diode detectors. There are benefits and limitations associated with
each type of sensor. Many of today’s average power measurements require a dynamic range greater
than 50 dB. Keysight’s approach to meeting this need is to create an average power sensor with a wide
dynamic range that incorporates diode stacks in place of single diodes to extend square-law operation to
higher power levels (at the expense of sensitivity).

Measuring Radar and EW Signals 184

Measuring power with an average power meter
An average power meter can be used to report average power and pulse power if the duty cycle of the
signal is known. There are some advantages to using this method but there are also several points that
must be taken into consideration. When an average power meter reports a pulse or peak power result,
it does so by deriving the result from the average power and a known duty cycle. The result is accurate
for an ideal or nearly ideal pulse signal but it does not reflect aberrations due to a nonsquare pulse shape
and will not detect peak excursions that may result from ringing or overshoot. The main advantage of
average power meters is that they are the lowest-cost solution. Both the power meters and sensors are
less expensive than the corresponding peak power meters and sensors. They also generally can measure
over a wider dynamic range, frequency range, and bandwidth, and can measure a signal no matter how
fast the rise time or narrow the pulse width.

Measuring power with a peak power meter

The peak power meter with a sensor has the advantage in that it is capable of making direct
measurements of peak power and pulse power. This is particularly useful for shaped or modulated pulses
for which deriving pulse power from average power may be inadequate.

Figure 17.3 shows an example of measuring peak and pulse power using the Keysight P-Series peak
power meter. A convenient feature of the meter is its trace display, which allows you to view the envelope
of the pulse signal that is being measured. The meter operates by continuously sampling the signal
with a 100 MS/s digitizer, buffering the data, and calculating the result. This gives the meter greater
measurement versatility, including flexible triggering, time gating with multiple gates, and the ability to take
single-shot measurements.

Figure 17.3. Using the P-Series peak power meter to measure peak power, gated pulse power, and peak-to-average
ratio. Due to the shape of this pulse the peak power is 1.39 dB higher than the pulse power.

Measuring Radar and EW Signals 185

Figure 17.4. P-Series power meters and power sensors

Keysight power meters operate with various sensors (CW, average, and peak and average) and cover
numerous frequency and power ranges to accurately measure the power of RF and microwave signals.
Keysight P-Series wideband power meters (30 MHz video bandwidth) such as the N1911A (single-
channel) and N1912A (dual-channel) provide measurements including peak, peak-to-average ratio,
average power, rise time, fall time, and pulse width.

When used with an N1921A or N1922A wideband power sensor, an N1911/12A P-Series power meter
provides a measurement frequency range of 50 MHz to 40 GHz.

Pulse Frequency and Timing Measurements with a Counter

Recently advanced counters with pulsed RF/microwave measurement capabilities have reappeared
in test and measurement product lines. These advanced counters provide a low-cost solution that is
easily configured for pulse frequency (PF), pulse width (PW), pulse repetition frequency (PRF), and pulse
repetition interval (PRI) measurements. Within a counter, the internal measurement engine is an advanced
event timer. Instead of digitizing a signal—as in an oscilloscope—the counter makes measurements by
triggering off a start event such as the rising or falling edge of a signal, at a specified amplitude. Once the
start trigger occurs, a timer is started and runs until a specified stop event trigger is reached.

Using a method known as interpolation, accurate timing measurements with sub-nano-second resolution
can easily be obtained.

Figure 17.5. Frequency counter

Measuring Radar and EW Signals 186

Measuring Pulse Power and Spectrum with a Signal/Spectrum Analyzer
The primary advantage of a signal analyzer is that it can measure the frequency content of the radar
in addition to power. This is important because an incorrect spectrum can indicate several problems
that result in wasted power and the emission of unintended signals. In general, an improper spectral
shape indicates a radar that is operating less than optimally. For example, Figure 17.6 shows radar
signal spectra before and after adjustment to the cross-field amplifier used for the radar transmitter. The
symmetry of the spectrum indicates an optimally performing radar.

Figure 17.6. A signal analyzer is useful for examining the shape and symmetry of the radar spectrum. This example
shows the spectrum trace before and after a timing adjustment in the magnetron.

Measuring pulsed radar with a signal analyzer is complicated by the different modes of operation
that occur, which depend on the resolution bandwidth (RBW) setting of the spectrum analyzer. These
variations exist when measuring any type of pulsed signal but tend to be more noteworthy when
measuring the low-duty-cycle pulses commonly used with radar signals. Further, different modes in signal
analyzers, namely swept versus fast Fourier transform (FFT), can behave differently when measuring
pulsed signals.

This section starts with a quick review of the basic spectral shape of a simple pulsed RF signal. Next, it
will examine the measurement of radar with both swept-based and FFT-based signal analyzers (including
different measurement modes) and then conclude with a survey of the different built-in measurement
functions included in many of today’s analyzers.

Measuring Radar and EW Signals 187

Figure 17.7. PXA series high-performance signal analyzers

Pulse spectrum review

In the time domain, multiplication of a continuous wave signal by a pulsed waveform results in a pulsed
carrier. The spectrum of a pulsed signal forms a characteristic sinc function shape with a main lobe and
sidelobes. From a mathematical standpoint, this can be understood by taking the Fourier transform of a
rectangular waveform and then translating it to the frequency of the carrier.

As can be seen in Figure 17.8, the pulse width and PRF of the signal determine the characteristics of the
basic pulsed spectrum. As the pulse width narrows, the width of the spectrum and sidelobes broadens.
The PRF of the pulsed-RF waveform determines the spacing between each spectral component. Viewing
the spectrum of the pulse can therefore provide meaningful information about the signal’s pulse width,
period, and duty cycle. For a basic pulsed-RF signal, the duty cycle can then be used to calculate the
peak pulse power from the average power and vice versa.

Period = T

PRF = 1/ T

V
~ Pulse width = 

Duty cycle =  / T

Pulsed carrier spectrum f

PRF = 1 1
T 
fc

Figure 17.8. Pulsed spectrum

Measuring Radar and EW Signals 188

Pulsed RF measurements with a swept spectrum analyzer
Conventional spectrum analyzers are based on analog super-heterodyne swept architectures. Most
modern instruments such as the Keysight PXA Series high performance signal analyzer or MXA midrange
signal analyzer employ a digital implementation of a swept architecture. This approach has many benefits
in speed and accuracy over their analog counterparts. Other spectrum analyzers may work by calculating
an FFT. Still others, including the PXA and MXA, employ both techniques. Each has its advantages. For
example, swept analyzers usually have the best dynamic range while FFT analyzers are likely faster for
computing in-channel measurements. Other differences, as they relate to measuring pulsed radar signals,
will be outlined below. One advantage of a swept architecture is that most RF designers are familiar with
its operation. That familiarity results in an intuitive understanding of signals from their swept spectrum
measurement that is lost in a snapshot FFT spectrum.

Pulsed RF measurements with analyzers that compute FFT

As mentioned above, some signal analyzers use FFT to compute spectra in a manner similar to that
of a VSA. Analyzers that use FFT techniques have advantages and disadvantages when compared to
swept analyzers. Swept analyzers have advantages in sensitivity and wide-span measurements. FFT
analyzers can be faster for measuring radars with bandwidths less than the analyzer’s maximum FFT
analysis bandwidth. FFT based signal/spectrum analyzers can also perform VSA measurements (if the
software is implemented) because phase information is maintained. However, for reasons covered below,
FFT analyzers tend to be inadequate for measuring wideband radar or radar with low duty cycles. Some
spectrum analyzers such as the PXA and MXA include both swept and FFT modes and automatically
switch between them depending on measurement settings. The analyzer can be set to automatically
optimize for speed or dynamic range or forced to remain in either FFT or swept mode.

FFT-based spectrum analyzers typically have a user interface designed to have a look and feel similar
to that of a traditional swept analyzer. A casual user may not even recognize that they are using FFT
techniques rather than traditional spectrum sweeps. However, the differences become apparent when
measuring pulsed RF signals.

Like a vector signal analyzer, the FFT-based spectrum analyzer can be fast at measuring signals whose
entire bandwidth is contained within a single FFT (i.e., contained within its analysis bandwidth). With this
condition met, the FFT signal analyzer is essentially equivalent to a VSA though typically without as many
measurement functions and displays.

When the span of interest is wider than the analysis or FFT bandwidth of the analyzer, the FFT-based
spectrum analyzer calculates the spectrum by taking multiple FFTs at different frequencies and
concatenating the results. This technique is sometimes called stitching because the analyzer computes
the spectrum one section at a time, step tuning to a different frequency for each section, and patching
them together. Depending on the speed of the analyzer, one may be able to see each section of the
spectrum appear as it is computed.

Measuring Radar and EW Signals 189

If the condition for line-spectrum mode is met (PRF < 0.3 RBW) then there is no difference in the result for
the traditional swept or FFT analyzer. However, if these conditions are not met, FFT-based analyzers will
behave quite differently than swept analyzers. In this case, the FFT analyzer will not display the PRF lines
seen in the swept analyzer. Rather, the data displayed will depend on the probability of intercept between
the FFT acquisitions and the pulses.

Built-in measurements
Today’s modern spectrum analyzers include many built-in functions and capabilities that can simplify and
enhance radar measurements. Several of these features are highlighted here.

Channel-power
The channel-power function is designed to measure the average power across a given frequency band. It
is a common measurement that is frequently used to measure many different types of signals. Spectrum
analyzers use different techniques for making channel-power measurements. The most common way,
which is usually the most accurate, is the integration bandwidth method. The analyzer essentially
integrates the power as it sweeps across the given integration bandwidth. Typically, the measurement
uses the analyzer’s averaging detector. Examples of channel-power measurements of a radar signal are
shown in Figure 17.9.

Chirped radar 7-bit Barker code

Figure 17.9. Channel power measurement of radar signals performed on a spectrum analyzer. Results are equivalent
to average power.

Measuring Radar and EW Signals 190

Occupied bandwidth
The occupied bandwidth (OBW) measurement automatically calculates the bandwidth in which a
specified percentage of the power is contained. The OBW of the signal is often determined based on the
bandwidth for which 99% of the signal’s power is contained, as shown in Figure 17.10.

Figure 17.10. This is an example of an occupied bandwidth measurement on the PXA spectrum analyzer.

Bandwidth for which 99.00% of the signal power is contained is automatically measured and reported.

Measuring Radar and EW Signals 191

Burst-power
The burst-power measurement is an automated zero-span measurement. Rather than integrating
the power in the frequency domain (as is done in the channel- power measurement) the burst-power
measurement integrates the power across a defined time slot or gate and is essentially equivalent to the
gated power measurements discussed earlier in the power meter section. The measurement often uses
a burst-power trigger (provided on some signal analyzers such as the PXA) that automatically finds and
triggers on the burst (or pulse), as shown in Figure 17.11. This can be very useful for radar because it is a
direct measurement of the pulse power. Its limitations, however, are the same as those for zero span: the
RBW filter must be wide relative to the occupied bandwidth of the signal.

Figure 17.11. This is an example of burst-power measurement from the PXA signal analyzer. Burst power and pulse
width are automatically measured in zero-span mode.

Measuring Radar and EW Signals 192

Measuring with a Vector Signal Analyzer
Unlike a spectrum analyzer, a vector signal analyzer captures the phase and magnitude information of the
measured signal and uses this information to perform more advanced analysis. Vector signal analyzers are
typically very flexible and can display results in the time, frequency, and modulation domains.

A vector signal analyzer does not sweep across a wide frequency range like a spectrum analyzer.
Most vector signal analyzers operate by tuning to a specific frequency, conditioning the signal,
down-converting, digitizing, and processing the signal. Some vector signal analyzers skip the analog
down-conversion stage and directly digitize the baseband, IF, or even RF signal after conditioning.

There are a variety of vector signal analysis solutions available with varying performance constraints and
tradeoffs in bandwidth, sensitivity, memory, and frequency range. Many of today’s modern instruments
include dual functionality and operate as both a spectrum analyzer and a vector signal analyzer, as is
the case with the Keysight X-series and PSA analyzers, or as an oscilloscope and vector signal analyzer.
Keysight VSA software can also be extended to work with logic analyzers to analyze signals in digital form
or to work in software simulation environments such as Keysight ADS or MATLAB.

Explore Your Signal Vector Mode

Even with basic VSA , one can learn much about their signal in Vector mode, which enables time domain
and frequency domain visualizations.

Figure 17.12 Time and frequency domain representation of chirp signal(left). Spectrogram view (right)

Measuring Radar and EW Signals 193

Radar Pulse Analysis
With the radar pulse analysis measurement extension, pulse boundaries are automatically detected
with advanced algorithms. A Swiss Army knife of tools becomes available, including new trace types,
statistics, measurements, and table metrics that are all specifically tailored for pulse analysis.

Trace Types - Here we show instantaneous

frequency and instantaneous phase versus time.
Furthermore, based on a best fit analysis of
instantaneous frequency, deviations from the best
fit are plotted as “FM Error vs Time.”

Statistics - Any metric that may be tabulated

in a pulse table may be analyzed in terms of its
statistics and trendlines. In this case, the amplitude
envelope overshoot is plotted as a histogram. This
way RF system engineers can answer the question,
“How accurate and repeatable were my pulses?”

Measurements - Certain workflows require

specialized measurements. Some examples
include the analysis of pulse sidelobes and pulse
compression; angle of arrival; or even pulse pattern
search. In this illustration, we present various pulse
trains that have been recognized and color-coded
appropriately.

Figure 17.13. Trace types, statistics and measurements

Measuring Radar and EW Signals 194

Figure 17.14. All-in-one view in single window of VSA software

All three elements (trace types, statistics, and measurements) are shown above with highly configurable
windows. Four pulses (4 through 7) are highlighted and aligned with their corresponding time-domain
traces. A histogram of the rise time, the trend line of the pulse fall time, and an overall pulse summary are
also included. Along the bottom, detailed pulse metrics are tabulated across the 40 pulses detected.

Measuring Radar and EW Signals 195

How Accurate and Repeatable were my Pulses?
Since pulses may span many gigahertz of frequency, large data sets must be analyzed to identify
individual pulses and quantify various figures of merit. With the option 89601BHQC, pulses are
automatically identified, labeled, and cataloged in pulse tables.

Figure 17.15. Pulse table

The power, frequency, amplitude, and phase vs. time are carefully analyzed for each pulse to calculate
figures of merit that can be tabulated in a pulse table (Figure 17.15). Example metrics are pulse width
(PW), pulse repetition interval (PRI), rise time, fall time, and amplitude overshoot illustrated below.

Figure 17.16. RF pulse output

Measuring Radar and EW Signals 196

Numerous other metrics are organized in the following categories:

• Modulation Metrics – modulation type, modulation code number, chip count, measured bits, chip
width, chip offset
• RF Output Level Metrics – top level, base level, top to base ratio, amplitude when the pulse is on,
peak level, mean level, peak to average ratio
• Amplitude Settling Metrics – pulse droop, droop rate, droop starting amplitude, droop ending
amplitude, overshoot, ripple
• Time Metrics – rise time, fall time, rising edge, falling edge, width, duty cycle, pulse repetition
frequency, pulse repetition interval, off time
• Frequency Metrics - mean frequency, pulse to first pulse frequency difference, peak-to-peak
frequency deviation, and relative to an estimated reference signal, RMS frequency error, peak
frequency error, and time location of peak frequency error
• Phase Metrics – mean phase, pulse to first pulse phase difference, peak to peak phase deviation,
and relative to an estimated reference signal, RMS phase error, peak phase error, and time location of
peak phase error
• Linear Frequency Modulation Metrics – best-fit mean modulation frequency, best-fit start modulation
frequency, best-fit ending modulation frequency, best-fit peak-to-peak modulation frequency
deviation, best-fit FM slope, integrated nonlinearity from best-fit
• Triangular Frequency Modulation Metrics – best-fit apex frequency, best-fit apex time
• Channel to Channel Difference Metrics – time, amplitude, and frequency difference of a
corresponding pulse on channel 2 as compared to Channel 1
• Pulse Compression Metrics – correlation between reference pulse and measured pulse; peak
sidelobe level, peak sidelobe location, compression ratio, main lobe width
• Deinterleaving Metrics – emitter ID
• Frequency Hopping Metrics – hop state index, hop begin time, hop ending time, hop settling time,
hop dwell time, hop switching time, hop mean frequency, hop mean frequency deviation

Any of these metrics may be copied to Microsoft Excel or exported to CSV format for further analysis.

Furthermore, statistics and trends on any of these metrics may be visualized. Hypothetically, a radar
engineer may have a large population of pulses that are supposed to have the same pulse width or pulse
repetition interval. The repeatability of the RF system may be plotted using a histogram.

Measuring Radar and EW Signals 197

Below we show a histogram of pulse repetition interval using an MXG signal generator. Judging by the
limits of the x-axis, we observe very little variation in PRI.

Figure 17.17. Pulse histogram

Evaluate Modulation on Pulse

A real-world signal might be hopping in frequency and modulated in different ways. We will need to
automatically identify frequency modulation and modulation code. Fortunately, the VSA enables automatic
pulse modulation recognition in the pulse results table, even providing decoded bits in some cases.

Figure 17.18. Modulation types

Measuring Radar and EW Signals 198

A small subset of the possible modulation types is shown in Figure 17.18. A more complete list includes
the following types: Continuous Wave, Linear FM, Triangular FM, Barker Phase, BPSK, QPSK, Frank
Code, P1 Code, P2 Code, P3 Code, and P4 Code. Furthermore, Bipolar Phase Shift Keying (BPSK)
modulation may be defined from 0° to an arbitrary phase, including the most typical 180°.

In larger test systems, trigger events may arrive intermittently, with significant pauses in between
trigger events. In such cases, it is helpful to tabulate pulse metrics across trigger events and individual
acquisitions. For these scenarios, the pulse table may be configured as a cumulative pulse table in that
rows describing groups of pulses may come from different acquisitions.

Characterize Angle of Arrival with Segmented Acquisitions

In situations where the duty cycle can be very small, much of the high precision and high bandwidth IQ
data captured is wasted. As an illustration, let’s assume a pulse width of 1 µs but with a pulse repetition
interval of 1 ms, implying a duty cycle of 0.1%. Graphically, this is not very informative as may be seen in
Figure 17.19.

Figure 17.19. Pulsed RF envelope

Measuring Radar and EW Signals 199

However, if we enable segmented capture, much of the dead time may be eliminated, allowing a closer
look at the most salient characteristics of the pulse, like overshoot or ringing.

Pulsed RF Envelope vs Segmented Time, Segment Lengh = 5μsec

0
1 ms 2 ms 3 ms 4 ms 5 ms 6 ms 7 ms
-10

-20
RF Envelope (dBm)

-30

-40

-50

-60

-70

Time (μs)
5 μsec 5 μsec 5 μsec

Figure 17.20. Segmented capture of pulse

By leveraging segmented capture measured across four oscilloscope channels, scan patterns across four
antennas may be characterized over tens of seconds as shown in Figure 17.21

Figure 17.21. Segmented capture across four channels

Measuring Radar and EW Signals 200

Once the antenna spacing and location have been specified, angle of arrival (AoA) calculations may be
completed using one of two methods: noting phase differences or the time difference of arrival (TDOA)
between channels. Below, a trend plot shows the evolution of elevation and azimuth across pulses
(Figure 17.22)

Figure 17.22. Trend plot showing AoA elevation and azimuth

Quantify Pulse Compression and Linear FM

Square Pulse

Binary Phase
Coded Waveform

Linear Frequency Mainlobe

Modulated Waveform Width Mainlobe
Peak
Matched Sidelobes Sidelobe
Filter
0 Time Delay
Figure 17.23. Pulse compression

Radar works by sending out a loud pulse and then listening for the echoes of that outbound pulse. In
other words, we judge the distance of an object by reflecting a signal off it. By noting the arrival time of
the echoes and accounting for the speed of propagation, we can estimate the distance or “range” of
the object. Furthermore, by noting the frequency shift of the echoes, we can also gauge the velocity.
By changing the shape of the outbound pulse in a specially designed way, the distance resolution and
immunity to ambient noise can be improved with the help of a matched filter. Thus the received signal
is submitted to a filter that is simply a time-reversed version of the original signal. This is the same as
taking the correlation of the original signal. Any echoes off real objects will have the same shape as the
outbound signal, and so the convolution of the received signal against the time-reversed outbound signal
leads to taller and narrower peaks, which is great for range resolution. System noise or radar clutter will
not have the same signature as the outbound signal, so after taking the correlation against the outbound
pulse, it continues to look like noise.

Measuring Radar and EW Signals 201

With the VSA, the matched filter is specified by choosing a pulse as a reference pulse and saving its
shape to a data register. Pulse compression may then be evaluated in terms of the agreement between
measured pulses and reference outbound pulse; sidelobe height and time offset; and main lobe width
as compared to the overall pulse width (or compression ratio). These are tabulated as metrics in the
pulse table.

By the 1950s, the industry had wholeheartedly embraced this concept of pulse compression, and people
were looking at improved ways of having sharper main lobes with lower sidelobes. Both experimentally
and theoretically, they learned that a wider bandwidth signal leads to lower sidelobes and taller main
lobes. The most common approach is to apply linear frequency modulation to the pulse. To achieve even
higher levels of pulse compression, radar engineers have explored “non-linear frequency modulation” or
NLFM, typically characterized as a polynomial fit to the frequency versus time data.

While some would argue one is a superset of the other, the VSA can analyze both types, returning the
polynomial fits to the instantaneous frequency versus time in the case of NLFM. Linear FM metrics
are tabulated in the pulse results summary table. For every pulse that has a linear FM chirp, the VSA
implements a best-fit to the measured frequency versus time data to estimate what would be the
instantaneous frequency versus time if we were looking at a perfectly linear modulator. Deviations from
this best-fit are then tabulated.

Getting Data into the VSA

What type of hardware instrumentation is well-suited for radar pulse analysis?

Many radar system engineers require tremendous instantaneous bandwidth because no one knows
precisely what frequency the pulses will be arriving. In such situations, we would recommend a
broadband oscilloscope with up to 110 GHz of instantaneous bandwidth. Whereas typically super-wide
bandwidth oscilloscopes have presented poor dynamic range with the full bandwidth of noise degrading
the measured signal-to-noise ratio, this trade-off is not necessary today. The Infiniium UXR scope features
hardware-accelerated decimation and filtering.

Some radar engineers are trying to find a relatively small signal hidden in the noise of a dynamic
spectrum. In that case, we would recommend a state-of-the-art spectrum analyzer like the UXA with
dedicated hardware to provide the best possible dynamic range.

Lastly, for a radar engineer on a test range interested in looking for intermittent aberrations in the
electromagnetic airspace, they will be interested in collecting huge amounts of data and saving it
to robust hard drives without any gaps between acquisitions. In that case, we would recommend a
streaming digitizer such as the M8131A with an optical interface to provide adequate data throughput.

The VSA connects to all three classes of instruments and provides a hardware extension to over 300
Keysight model numbers allowing an engineer to test anywhere in the transmitter chain, from the DSP
with a logic analyzer, to the baseband inputs with an oscilloscope, and the RF output with a signal
analyzer. Once a connection has been established, the software feels like an extension of the instrument,

Measuring Radar and EW Signals 202

allowing the user to set center frequency, sample rate, trigger on RF bursts of power, and even record
acquisitions to a binary file for post-processing. Furthermore, connectivity to third-party receivers may be
developed using the 89630B Software Development Kit (SDK) and then leveraged with the 89601300C
option. The most typical method for getting data into the VSA is by way of a hardware connection.

Figure 17.24. Hardware connectivity with 300+ models

A second option is through simulation results. The VSA software connects to both PathWave Circuit
Design (formerly known as Keysight Advanced Design Systems) and PathWave System Level Design
(formerly known as SystemVue). The third option is through recording files. Assuming the sample rate and
center frequency are well-known, waveforms may be brought into the VSA that may have been saved to
disk in Matlab format or various other binary formats.

Record Signals as IQ Waveforms or Pulse Descriptor Words for RF Playback

The VSA may be used to record long-time records as a way of monitoring trends or capturing unknown
signals, irregular events or new threats yet to be understood.

A radar engineer can save measured RF data to an IQ data recording and then use that same data
recording to play it back through a vector signal generator. The I and Q outputs from the data recording
get played out as pulsed waveforms at RF frequencies.

With long-time recordings leading to many gigabytes of data, the radar industry invented this concept of
Pulse Descriptor Words (PDW’s), which allow an entire pulse to be submitted as a series of parameters
describing that pulse. The most relevant characteristics like pulse width, PRI, and mean frequency are
submitted as columns to a PDW file, where every row represents a different pulse. This PDW file may be
then uploaded to an agile vector signal generator like the UXG.

Whether through an IQ data recording or using a PDW file, the recorded signal may be played into a radar
receiver under test or radar jammer as a way of verifying its response or in EW parlance, verifying its

Measuring Radar and EW Signals 203

“techniques.” A radar analyst may thus capture signals from the sky or a test range and then submit that
same captured signal to a radar receiver under test.

Characterize Individual Pulse and Pulse Train Similarity

An RF system engineer typically knows what the outbound pulses should look like. Their job is to
generate accurate pulses and then look for impairments caused by the realities of the actual hardware.
If they find a badly formed pulse, they will need to understand why and how to fix the component or
subsystem responsible. On the other hand, a radar analyst typically collects large volumes of RF data
on a test range and checks whether the overall system responded appropriately to an external threat.
Typical questions might include: How long did it take for the emitter to change modes? Or, did the radar
switch from search mode to track mode at the correct time? On the electronic battlefield, an adversarial
signal may be trying to jam or confuse your radar receiver. To avoid this, the radar analyst maintains a
catalog of pulse patterns as well as pulse characteristics identifying the equipment responsible. Measured
pulses are thus compared against a catalog of known pulse trains and equipment signatures. To address
the divergent requirements of both the RF system engineer and the radar analyst, the VSA provides two
features - pulse scoring and pulse train search.

In both cases, we begin with the definition of some reference pulses. This is entered as a table with
characteristics of amplitude, pulse width, pulse repetition interval, and average frequency. Whereas pulse
scoring compares each corresponding pulse against the reference list of pulses, pulse train scoring
evaluates the agreement of the entire measured pulse train. With pulse scoring, one may answer the
question, “Is a pulse or series of pulses similar to a single reference pulse or series of reference pulses?”
On the other hand, with pulse train scoring one may answer the question, “How do we know that pulses
measured in a reference train happened in our latest measurement, and how closely do they match?”
Certain figures of merit (like pulse width) may be emphasized more strongly than the others by way of
base error tolerances. If the base error tolerances are set small, then that metric will play a stronger role
in the overall score. Once the reference pulses and base error tolerances have been defined, measured
pulses are color-coded both on the time trace as well as in the pulse summary table, as shown below.

Figure 17.25. Pulse train

Measuring Radar and EW Signals 204

Fast-Time Slow-Time Display and Doppler Plots

C B A

SUT C B A

SUT C B A

Figure 17.26. Input-to-output difference

A System Under Test (SUT) will receive a train of pulses and then in response, output a train of pulses.
To characterize the efficacy of this response, it becomes necessary to chart the pulses at the input and
the output of the SUT. A multichannel scope, like the UXR, may be used to collect phase-coherent time-
domain data to look at channel-to-channel differences or more specifically, input to output differences.
Based on the number of channels in the receiver, multiple SUT’s may be tested in parallel.

Radar engineers often want to track the evolution of the SUTs response to an incoming pulse, so the
pulse repetition interval demarcates a “fast-time” slice. In plotting the evolution of the SUT response
across fast-time intervals, a new notion of time lends itself to the name “slow-time.” Below, we show the
incoming signal on channel 1 and SUT response on channel 2 in Time vs. Time Heatmap traces. This
illustrates a classic “Range Gate Pull-off” technique.

Figure 17.27. Freq vs Time heatmap

Measuring Radar and EW Signals 205

A frequency-domain representation of the output signal can reveal a Doppler shift as evidenced by
slight frequency shifts in the SUT output signal. To capture time-domain data with high fidelity, a wide
bandwidth is required. However, this wide bandwidth provides a macro view of the frequency spectrum,
preventing a proper view of the frequency shifts due to the velocity change of the target.

The span needs to be narrowed significantly to support the most likely frequency shifts. After changing
the span, VSA resamples the same data to show the changing frequency impulses versus time, with the
most recent time slice displayed at the bottom.

This however results in a significant blurring of the details in the Time vs. Time Heatmap. So how can we
achieve good resolution in the time-domain as well as in the frequency-domain? The multi-measurement
capability in the VSA enables this, where the span for Meas01 may be set to 16 MHz for better Time vs.
Time resolution and independently, the span for Meas02 may be set to 100 kHz for better Frequency vs.
Time resolution. Both the Time vs. Time Heatmap and Frequency vs. Time Heatmap traces are offered in
basic vector mode and pulse analysis mode.

Since the days of Nikola Tesla and other pioneers, radar has become ubiquitous and the breadth of its
application is still growing. At the same time, radar technology has become more sophisticated as signal
processing is commonly used to enhance the returned signal and to extract information, such as target
images during post processing. However, no matter how sophisticated the signal processing becomes,
the performance of every radar and EW system is directly determined by the quality of the underlying
transmitters and receivers.

Measuring Radar and EW Signals 206

Understanding radar measurements and how instrumentation responds to radar signals is crucial to
designing high-performance and cost-effective radar solutions. This chapter has sought to provide an
explanation of common radar measurements and highlight the measurement solutions available today.
Critical radar measurements include power, spectrum, pulse characteristics, antenna gain, target cross-
section, component gains and losses, noise figure, and phase noise—all of which can be shown, through
the radar range equation, to directly influence the performance of the radar. These measurements
should always be made using quality test equipment designed to deal with the unique and demanding
characteristics of radar systems and signals.

WIDEBAND, MULTI-PORT, COHERENT RF PULSE ANALYSIS

https://www.keysight.com/see/radar-ew-subsystem-test

Measuring Radar and EW Signals 207

Chapter 18
Testing TR Modules Efficiently
Radar, satellite, and electronic warfare (EW) systems utilize a wide variety of microwave modules.
Active Electronically Steered Arrays (AESA) is one of the most critical and complex modules of a
Synthetic Aperture Radar (SAR).

A phased array antenna with an array of antenna elements controls the relative amplitude and
phase of signals from these elements to concentrate the energy in the desired direction. The
attenuator and phase shifter in each TR module quickly change their states, which enables a
phased array antenna to scan the narrow beam very rapidly.

Testing TR Modules Efficiently 208

 Wave front
Delay

Beamforming Network (BFN)

Phase shifters and attenuators

Figure 18.1. AESA beamforming network

Today’s phased-array antennas utilize 100s or even 1000s of antenna elements with TR modules. Their
energy can be directed and scanned to multiple positions, which makes them extremely flexible for use in
radar, electric warfare, and communication applications.

Active electronically scanned arrays, or AESA antennas, are the latest version of the phase-array antenna.
AESA can form a lot of antenna patterns. This provides versatility with a wider scan angle and multiple
concentrated beams for detecting and tracking multiple targets simultaneously.

Precise relative phase and amplitude control are crucial to the AESA capabilities. Shown here is a
typical AESA antenna with 24 test ports and 6-bit phase and 6-bit attenuator control. It has 4096
settings per test port and is tested at 21 frequency points. That is over 2 million antenna patterns to be
tested, Figure 18.2.

4096 antenna patterns per frequency and test port

Θscan angel: ±45º
Φscan angle: ±45º
24 test ports
21 frequency points
4096 ‘beam’ positions per
port and frequency

24 x 21 x 4096 =
1
4096 2 million antenna patterns!

Beamforming network

Figure 18.2. AESA test

Testing TR Modules Efficiently 209

Table 18.1 shows three types of phased-array antenna systems with their typical applications and
characteristics.

Type 1 Type 2 Type 3

Communications, Communication on Fire Control, EW, ECM,

weather radar, ships, airborne etc.
perimeter detection, intelligence,
imaging radars, surveillance,
automotive radars, etc. reconnaissance (ISR)
radars, etc.

30 to 100 mW per Medium power, can be High-power, 3 to 10+W

element on Tx, can be done in SiGe + per element, GaN,
done in SiGe + lowpower mediumpower GaAs maybe SiGe
GaAs

$ $$ $$$$$

Table 18.1. AESA-applications and characteristics.

Source: TRM tutorial session at IEEE International Symposium on Phased Array Systems & Technology; October
17-21, 2016.

Type-1 RADARs are low power and low-cost arrays for communications, weather radars, perimeter
detection, and automotive radars. The range of applications is rapidly growing. The phased-arrays
needed for future 5G systems will be in this category. Type-2 RADARs are medium power phased-arrays
used for military intelligence and surveillance which have a higher cost per element. This type of radar
system is also growing rapidly as communication and situation awareness become more and more
important. Type-3 RADARs are very high power per element and very expensive. These are used in
applications for fire control radars, electronic warfare, and electronic countermeasures. GaN is the current
semiconductor of choice for these high-power applications.

Applications for phased-array antennas are expanding and growing rapidly in aerospace and defense to
commercial industries, with medium to low power but highly integrated form factor. TR modules typically
need extensive testing to ensure that they are matched across the phased-array in which they are used.

Testing TR Modules Efficiently 210

 Digital Beamforming Analog Beamforming
• Signal generator and receiver per element • Shared signal generator and receiver
• Minimum power loss and increased • Larger power and receiver sensitivity
receiver sensitivity losses
• Most flexible, can scan with whole, subset, • Limited flexibility
or each element • Simplicity enables high-density in small
• Complicated, requires space and power space
ADC
DAC
Hybrid Architecture
ADC
DAC
DAC

Beamforming data processing

ADC
DAC ADC
ADC ADC
DAC
DAC
DAC Σ
ADC
DAC ADC
ADC
DAC DAC
ADC
DAC
ADC
ADC
DAC
DAC

6
ADC

Table 19.2. Digital vs Analog beamforming

Digital beamforming in phased-array antennas is becoming popular in both military and commercial
applications. It has a radar exciter, or signal generator and a receiver right behind each TR module,
and provides the most flexible beam control. This configuration minimizes power loss in the transmitter
and increases receiver sensitivity. The tradeoff is that it increases the complexity and requires more
components and space than analog beamforming. An alternative approach is a combination of analog
and digital beamforming to optimize special and power requirements, as well as performance and
versatility requirements.

The active phased array allowed lower RF power per element in the front-end. Lighter and smaller
structures became replacements for bulky waveguide manifolds. This enabled more elements in a system
and allowed more sophisticated digital beamforming.

Then with the planar structure, the whole system became smaller, lighter, and cheaper, but required more
components to be integrated into a smaller form factor.

Now, as function blocks like front-end TR modules, signal generators, and receivers baseband digital
processors become available in IC chips and frequency range is extended; the phase array antenna
systems will find applications in many industries. Then we begin to see needs of different scales and
sizes, lower-cost structures, high-volume production, and maybe new performance requirements.

Testing TR Modules Efficiently 211

Ultimately, front and back-end RF and digital sections will be fully integrated and attached directly to the
antenna array. It will drive full scalability in size, cost, power, frequency, and bandwidth making it easier to
adapt to different requirements.

Trends and Challenges

TR modules have a major effect on RF performance. During transmit operations the output RF pulse
is amplified by the module, thereby defining the maximum radiated power of the radar. Because the
transmitter is operated in pulsed mode, output pulse parameters are typically measured. During receive
operations, the low-noise amplifier (LNA) within the module input determines the system noise figure
and consequently the minimum detectable signal. Within each path, programmable phase shifters and
attenuators control the antenna beam-steering and determine the angular accuracy of the radar.

Now, let’s review the front-end TR module and its trend.

Here we see a typical TR module. In one direction it’s a pulsed power amplifier, in the other, it’s a low
noise amplifier. All the RF characteristics of these paths are dominated by these devices and circuit
designs around them.

TR Module Air-Interface

Amp PA

Phase
Adjust
TRM UUT
Ant
RF RF
TRM Fixture
Atten
Adjust
3.5mm 3.5mm
Power & Control
Air interface
LNA TRM UUT RF
Tx/Rx RF
Control TRM Fixture Waveguide
Coupler
3.5mm 3.5mm
Power & Control

Figure 18.3. TR module

The unique part of a TR module is the phase shifter and attenuation adjusters which allow each TR
module to have its own settings. The attenuator and phase states can be re-programmed for each
radar pulse.

When TR modules are tested, they are attached to a fixture with a coaxial interface or waveguide adapter
for modules with built-in antennas. Then, the fixtures are de-embedded from the measurement results.

Let’s now focus on TR modules most commonly used today and discuss their test requirements
and challenges.

Testing TR Modules Efficiently 212

Figure 18.4. TRM transmit and receive path

Let’s start with the transmit path (Figure 18.4). The signal in the transmit path is typically pulse-modulated.
It goes through an attenuator and a phase shifter, and then a power amplifier. To deliver amplitude-
and phase-controlled signals to the antenna element, the input and output matches of the path, gain,
and attenuation and phase offset deviations are measured at every possible setting over the operating
frequency range. The attenuation step can be 0.5 or 1 dB in the 20 to 30 dB range, and the phase offset
is controlled in about 3-degree steps or less from -180 to +180 degrees. The number of combinations, in
this case, varies from a few to several thousand. Each set of measurements are simple S-parameters, but
there are a lot of amplitude and phase states to be tested.

Furthermore, the power amplifier characteristics are tested. This includes a 1 dB compression point or
P1dB, maximum output power at P1dB or at saturation, 2nd and 3rd harmonic distortions, and output-
referred IP3. It is quite simple to measure the gain, then find P1dB and the output power with a vector
network analyzer. Notice they are all tested versus frequency. If it is tested at every 10 MHz in the whole
X-band, the measurements have to be repeated 401 times.

Distortions are traditionally tested with signal generators with a combiner and a spectrum analyzer.
This approach is simple to implement and very straight-forward for fixed-input frequency, but it is time-
consuming for testing versus frequency.

Now, let’s look at the receive path. The receive path has a power limiter at the front-end to protect the
following low-noise amplifier from unexpected high-power signals. The phase shifter and attenuator are
controlled in the same manner as the transmit path. The matches and gain are tested over the operating
frequency and the various attenuator settings. They are simple measurements but require a lot of steps.

The input path is the front-end of the whole receiver chain where distortions are tested. They include 2nd
and 3rd harmonics, input-IP2 and IP3 or IIP2 and IIP3. They are calculated with the main output power,
2nd- and 3rd-order IMD, and the gain. A spectrum analyzer is used for distortion measurements, and a
VNA measures the gain. Testing IIP2 and IIP3 requires data from both instruments and some math. They
are also tested versus frequency, so it becomes even more complex.

This is probably not common in production, but TR modules may be tested with transmitter and receiver
subsystems in design characterization. The matches stay simple as S11 and S22, but transmissions
become conversion gains between the input and the output at different frequencies. Amplitude

Testing TR Modules Efficiently 213

measurements are relatively simple, although calibrations require extra steps. Phase measurements need
a slightly different measurement and calibration technique.

As there are frequency converters in the paths, there are more opportunities for signals leaking through
undesired paths. High-order mixing products are generated so there are more spurs. The mismatches
in the paths also make the signal flow more complex. These spurs and leakages must be tested in
transceiver subsystems. Additionally, wideband performance with modulated signals may be tested,
such as noise-power ratio, transmitter noise density, and receiver error vector magnitude. Vector signal
generators and signal analyzers may be used for testing these performances. They may not be needed
in a TR module test system, but it is better to be able to incorporate these additional capabilities
when needed.

TRM Test System

Having understood the trends and TRM measurement challenges, let’s understand considerations for
building TR module test systems for higher test throughput and better test quality.

Data saving/analysis/reporting
Operator interface

Measurement parameters
DUT setup ranges/steps
Test sequencer

DUT interface User

System Test fixture
Test executive supplied
verification
hardware
and
Switch path Switch matrix and
diagnostics Pin layouts
control Signal conditioning software
Protocol
(PAs, LNAs, Filters, Attenuators, Isolators)
Measurement
application
Vector network analyzer Power meter
Instrument Signal analyzer Oscilloscope Digital I/O
firmware Noise figure analyzer DC power supply DC power supply
Signal generator Multimeter

Figure 18.5. Conceptual building block for test system

Figure 18.5 shows a conceptual building block for test system integration. Building a test system
involves three key areas of challenges viz. measurements, switch matrices and signal conditioning,
and automation.

The measurement challenges are mostly addressed with tools and techniques that are available from
instrument suppliers. Users need skills and knowledge to utilize them, but suppliers are typically able
to assist.

Testing TR Modules Efficiently 214

The switch matrices and signal conditioning are unique to each test system and measurement
requirement. It requires a good understanding of test specifications, RF and MW, and perhaps digital
fundamentals. Suppliers for switches and signal conditioning devices will be able to identify the right
products based on your requests, but you could spend a great amount of effort designing and optimizing
the configuration.

Automation is a huge task. Controlling TR modules and sequencing the tests require a deep
understanding of the module design, test plans, specifications, and expertise on hardware control and
software integration. It also requires measurement knowledge to maintain the test quality so that the test
systems perform as expected.

Before understanding the latest measurement techniques, let’s review the traditional test approach.

Traditional test systems are configured with many instruments, each of which was designed for a specific
set of measurements. For example, a power meter is used for output power measurements and a VNA
is used solely for S-parameter measurements. Separate signal generators and spectrum analyzers are
used for distortions and spurious tests. A noise figure analyzer is added for the NF measurements on the
receive path.

Here is an example of a VNA with the latest hardware architecture (Figure18.6). This is PNA-X with a
2-port configuration. It can have two internal sources with built-in harmonic filters with a signal combiner
for two-tone IMD measurements.
+28V J11 J10 J9 J8 J7 rear panel J2 J1

+
- LO
Source 2
OUT 1 OUT 2 Noise receivers
Source 1 To receivers
Pulse
OUT 1 OUT 2 modulator 10 MHz - 3-
3 GHz 26.5
GHz
R1
Pulse
modulator Pulse generators
R2
A 1
2
B
3
4

Source 2 Source 2
Test port 1 Output 1 Output 2 Test port 2

DUT
Impedance tuner for noise
figure measurements

Figure 18.6. VNA architecture

Testing TR Modules Efficiently 215

The sources are capable of pulse modulation, controlled by internal pulse generators, which is useful
for testing devices in radar applications. There are signal path switches behind the reflectometer so that
signals are routed to different hardware for additional measurements while maintaining S-parameter
measurement accuracy. Additional hardware may be included, such as a low-noise receiver for noise
figure measurements.

The test fixture is connected to different instruments through a large multiplexing switch matrix, which
adds large path loss and mismatches, causing drift and reducing system dynamic range. Software must
control each instrument with a slow interface and longer instrument response time. This becomes a major
problem for sequencing a lot of tests like TR modules.

These measurements are traditionally made with standard S-parameter measurements in a VNA. The
compression, or P1dB, is found in swept power S21 measurements then the output power can be found
at the same output power as the P1dB. It needs many setups to cover the required frequency range
which causes throughput and calibration challenges.

The latest VNA, like the PNA-X, offers a software option called Gain Compression Application or GCA.
In a single setup, the GCA controls the stimulus frequency, power, linear gain, P1dB, and output power
at P1dB versus frequency with fewer measurement points. Using the GCA, the setup and calibration are
simpler and the measurement speed is significantly faster than the traditional swept-power S21 method.

Spurious tests are typically done using a spectrum analyzer with fixed-frequency input to the DUT. The
input frequency is stepped to cover the frequency range. The search range becomes wider frequency and
lower level making it time-consuming. Repeating these measurements at different input frequencies can
lead to hours of testing.

The slow sweep speed is mostly due to the microwave preselector at the spectrum analyzer input in
swept mode. Spectrum analyzers use it for measurements with the highest level of accuracy. The latest
spectrum analyzers also have FFT mode for faster measurements, but it bypasses the pre-selector and
loses accuracy.

Let’s move on to IP2 and IP3 measurements. Today we have a simpler way with swept-IMD applications.
The swept-IMD controls the two internal sources and switches to make two-tone IMD measurements.
By simply setting the center frequency, tone-spacing, and the sweep range we can choose the test
parameter from the pre-defined list. This is easier than setting up individual source frequencies,
programming to sweep, finding peaks, and calculating parameters using the traditional method.

One of the most important parameters of the receive path is the noise figure, which is defined as signal-
to-noise ratio degradation. The noise figure is traditionally measured with either a noise figure analyzer
or a spectrum analyzer with a noise figure option using a technique called the Y-factor method. The
Y-factor method calculates noise figures from noise power measurements at the DUT output with hot
and cold noise source states at the DUT input. This is very popular and works well when the noise
source is connected at the DUT input. However, it assumes the DUT input is terminated with 50-ohm and

Testing TR Modules Efficiently 216

calculates the noise figure. When configured with switch matrices in an ATE system, this is no longer true,
and the noise figure is greatly affected by the input mismatch. This method is also very slow, which limits
the number of frequency points in practical use.

The noise figure application on PNA-X uses an optional low-noise receiver or standard VNA receivers
that minimize the need for switching between the noise figure and other measurements. With industry-
unique source-mismatch error correction, it delivers superior accuracy, especially in an ATE environment.
It calculates noise figure from the device gain and noise power at the DUT output with no signal at the
DUT input, so it is called the “cold-source” method. This method enables typically 10 to 40 times faster
measurement speed than the traditional Y-factor method, enabling a finer frequency resolution with a
small impact on the test throughput.

While this chapter focuses on the PNA series VNA for the TRM test, a similar but slightly modified approach
can also be used with Keysight modular PXIe VNAs and latest the E5080B ENA for TR module testing.

Classic ATE systems are configured with front-end switch matrices to connect multiple instruments to a
DUT. The design is simple, and it is easy to understand, but it is large, difficult to configure, less accurate,
and very slow.

The ATE systems with the latest approach introduce more software capabilities and utilize front-end
receivers for multiple measurements. However, not all required measurements can be done with a single
type of receiver using software. You may need a signal generator for higher output power or complex
modulation or a signal analyzer for wideband demodulation. Then you need to switch the paths to these
instruments. And you would most likely need signal conditioning for some measurements and switch
them in and out when changing measurement types. The best solution is to keep the front-end receivers
as close as possible to the DUT interface so you can minimize the path loss and system drift. This is not
always possible, but it is a good practice to optimize the path switches and maintain higher accuracy.

Testing TR Modules Efficiently 217

These signal conditioning devices often need to be switched in and out to optimize the performance
of each measurement. A high-power setup for the transmit path is completely different from the one
for noise figure tests on the receive path. We discussed earlier that switching should be minimized
between DUTs and receivers, but there are many cases where you must switch them in front for better
measurements, efficiency, or simplicity. If you can’t avoid the switches, they must be linear, very low loss,
and a good match. For these reasons, electro-mechanical switches are commonly used. There are a few
different types of switches with unique capabilities (Table 18.3).

Multiport switch Bypass switch Transfer switch

- One input 2 to 6 ports - Configured with a pair - Configured with a pair of
outputs of 1x2 switches 1x2 switches
- Used to switch multiple - Alternate signal paths - Switch to opposite states to
DUTs to one instrument, • Switch in/out a PA/ • Switch between 2 inputs
or multiple instruments isolator at the DUT input and 2 outputs
to one DUT • Switch LNA and attenuator • Used for signal reversal
path for a receiver • Configure as SPDT or
bypass switch

Table 18.3. Building switch matrices for signal routing

The first type of switch is a multiport switch. It has one input and typically from 2 to 6 output ports and is
used to expand the number of test ports for multiple DUTs, to route a signal generator to multiple outputs,
or sharing a signal analyzer with multiple input ports.

The second type of switch is a bypass switch. It is configured with a pair of one-by-two switches and
used for switching a signal path from one to another. For example, it is used to switch in and out a
booster amplifier and an isolator pair at the DUT input or used in front of a receiver and switches between
a high-sensitivity path with an LNA and a high-power path with an attenuator.

The third type of switch is a transfer switch. It is also configured with a pair of one-by-two switches, but
they switch to opposite states to switch between two inputs and two outputs and reverse the signal
direction. It can be configured as a single-pole, double-throw, or bypass switch. It is often used as a
bypass switch and for reversing the DUT connections between transmit and receive paths to simplify the
test system configuration.

Testing TR Modules Efficiently 218

Once you have designed and built a test system with switch matrices, the next challenge is to design
a calibration procedure that maintains the system’s accuracy and minimizes user effort and errors.
One approach is to integrate high-performance switches and calibration standards, such as an ECal
module, a power sensor, and perhaps a noise source and a thru path. This can be expanded to multiport
DUT topology with multiport switches. All paths from calibration standards to the reference planes are
measured and stored as user characterization files in the ECal module and the PNA. This approach
simplifies the calibration procedure and minimizes operator errors.

TR Modules need to be tested under environmental conditions. Under long temperature cycle for radar
and EW applications and TVAC conditions for satellite applications.

Shown in Figure 18.7 is a configuration example with CalPods for measuring TR modules in a thermal
vacuum chamber.

CalPod
Switch Matrix Switch Matrix
T/R module

T/R module

T/R module

T/R module

T/R module

T/R module

T/R module

T/R module

CalPod Controller

PNA-X
Figure 18.7. Calpod for maintaining measurement accuracy

Testing TR Modules Efficiently 219

The CalPod assemblies include multiple impedance states that are characterized over temperature range.
They are left in-line during the calibration and measurements and allow re-correcting the drift or any path
characteristic changes after the initial calibration. This approach increases calibration intervals, and saves
a lot of operation cost, especially for testing devices in thermal vacuum chambers.

It is one of the most critical aspects of the TRM Test system to synchronize TR module attenuator and
phase shift and getting the TR module state to change at every pulse of the radar’s PRF. You can see the
trigger handshake interactions between the PNA-X and the DUT control FPGA.

The PNA-X provides very flexible and low-latency triggering interfaces with DUT control FPGA, as well as
other instruments in the system.

The DUT control FPGA manages the timing of measurements, detects the measurement completions,
and attenuator and phase shifter states of the TR module over the proprietary interface.

Keysight’s Pathwave Test Automation Platform is a Microsoft .NET-based framework, designed for speed
and optimized execution. It provides a graphical user interface, which allows the user to quickly construct
test plans with highly repetitive tasks. Customizable modular plug-ins are available for test steps,
instrument/DUT interfaces, and result storage.

Pathwave Test Automation connection manager helps easily control the DUT, PNA-X, and non-Keysight
instruments, and a variety of measurements, including switch matrices. Pathwave Test Automation
command-line interface enables integration with other manufacturing applications and allows for various
levels of customization. There are many plug-in tools for visualization and analysis as well.

For TR module testing, Keysight provides a custom PNA-X driver which can be scaled to execute
a unique set of measurements and be synchronized to the DUT control required for phase shift and
amplitude validation. The results can be shown in test development mode or automatic test mode
in graphical format. The timing analyzer shows each test duration and provides insight into where to
optimize the test sequence for the maximum throughput.

Pathwave test automation plug-ins can be used to export the data into a graphical or numerical format for
easy analysis, test reports, and data archiving. The example here shows a graphical display on the test
automation platform as PNA-X will display. All data can be saved with a variety of standard data formats.

Keysight Pathwave test automation is not just another programming language, it is a highly customizable
and adaptable platform to speed up system integration for any skill level. It builds environments to archive
plug-ins and re-uses them to make future system integration simpler and quicker.

Testing TR Modules Efficiently 220

Keysight leads the industry in the development of new and advanced measurement methodologies
and techniques. TRM Automated Test solutions brings leadership experience to unique test needs and
provides a standard configurable platform giving the ease of use and flexibility required to reduce TR
Module test times without compromising measurement accuracy.

MEASURING T/R MODULES WITH KEYSIGHT PNA-X

COURTESY OF MICROWAVE JOURNAL
https://www.keysight.com/see/radar-ew-subsystem-test

Testing TR Modules Efficiently 221

Chapter 19
Understanding the Techniques for Antenna Testing
We work in a high-tech industry, where technology constantly changes and improves the way of
doing things. To remain competitive in the radar and EW industry, we need to continuously evolve
how we test antennas and RCS.

Before the mid-1980s, antenna and radar cross-section (RCS) test engineers used dedicated
microwave receivers. In the mid-1980s, utilizing a network analyzer in an antenna or RCS
application was a new and novel idea. Companies and individuals who adopted the new network
analyzer technology to make antenna or RCS measurements were leading innovators, and many
others came to follow this technology lead in later years.

With the next generation of network analyzers now available to the industry, the antenna test
community needs to evaluate this new technology to see if it can provide similar gains in improved
performance, accuracy, and speed to provide a better value for the antenna test community.
Antenna Test Parameters
Normally the antenna is tested under receive mode. However, certain applications require the antenna
to be tested in transmit mode also. Numerous parameters exist for characterizing the performance of
antennas and the most significant of these are discussed here.

• Antenna Impedance
• Radiation Pattern
• 3dB Beamwidth
• Directivity
• Antenna Gain
• Bandwidth

The antenna pattern is the response of the antenna to a plane wave incident from a given direction or the
relative power density of the wave transmitted by the antenna in each direction. The antenna pattern can
be realized by one of the three antenna pattern measurement techniques. The first technique developed
was the far-field range, then near-field techniques, and compact range.

Near and Far-Field Regions

Let’s understand the antenna radiating regions. It is divided into three regions:

Radiating Near-Field
(Fresnel Zone) Radiating
Radiating Reactive Far-Field
Antenna Field (Fraunhofer
Zone)

2 D²
4 2D²

rZ 2D² or_rZ10

Figure 19.1. Antenna radiating regions

1. Reactive: This region is the space immediately in front of the radiating portion of the antenna. The
reactive region is from the surface of the antenna up to one wavelength away from the surface of the
antenna. Successful measurements cannot be made in this region.

Understanding the Techniques for Antenna Testing 223

 2. Near-field: Beyond the reactive region, is the near-field region. The near-field region is often thought
of as existing from one to ten wavelengths from the aperture of the antenna. Because of geometric
scan considerations, the near-field is usually sampled at one wavelength from the antenna aperture
but can be sampled successfully at up to ten wavelengths.
3. Far-field: The transition from the near-field region to the far-field region is gradual, and not well
defined. The far-field region has been historically defined as 2D2/λ. This distance provides for
22.5°of phase taper across the aperture (as defined by ‘D’) of the antenna. For low-performance
antennas, this 22.5°of phase taper provided acceptable errors in the nulls and sidelobes of the
antenna. However, the actual far-field distance that is required is more of a question of the amount
of measurement error one is willing to accept in the null depths and sidelobes. When trying to
accurately measure a very deep mono-pulse null or a very low sidelobe antenna, one may find
that 10D2/λ may be required to reach the far-field criteria necessary to achieve adequate
measurement results.

The far-field range was the original antenna measurement technique and consists of placing the AUT a
long distance away from the transmit antenna. The AUT is illuminated by a source antenna at a distance
far enough to create a planar phase front over the electrical aperture of the AUT. The AUT moves Azimuth
and Elevation directions when measuring the antenna pattern. The far-field technique can be implemented
outdoors or indoors with a reflector (compact range).

Far-field antenna ranges were the original technique for characterizing antenna radiation patterns and
have been in use for over 60 years. However, as the antennas have become larger and have increased
performance, the far-field range distance has increased. Many factors have affected the viability of using
longer far-field antenna ranges. Longer antenna ranges are subject to increased undesired reflections
from the ground, buildings, and other man-made structures. Electromagnetic congestion has increased
the interference from other sources, as well as the need to reduce electromagnetic interference from the
antenna test range. All these factors have resulted in the need to find an alternative to far-field testing.

The first alternative to far-field testing is near-field testing, which is a technique that has been known for
many years but did not come into popular acceptance until the availability of adequate computational
power of computers. The near-field technique measures amplitude and phase data at half-wavelength
intervals across the radiating aperture of an antenna and utilizes a two-dimensional Fourier transform to
transform this measured near-field data to an equivalent far-field radiation pattern. Near-field techniques
began to gain popularity about 20 years ago, as the computational power of PCs was able to handle the
two-dimensional Fourier transforms. Today, near-field measurements are widely used by antenna test
engineers because of their many benefits. They utilize smaller carbon footprints, are not as affected by
electromagnetic interference, nor do they contribute to electromagnetic interference, provide all-weather
testing capabilities, provide security for testing proprietary antennas, and the errors associated with near-
field antenna ranges are much smaller and better characterized than for far-field antenna ranges.

Understanding the Techniques for Antenna Testing 224

 Planar Cylindrical Spherical

Figure 19.2. Scan type for near field antenna testing

There are three types of near-field scans:

• Planar Near-field scanning: The measurements are conducted by scanning a small probe antenna
over a planar surface. These measurements are then transformed to the far-field by use of a Fourier
transform.
• Cylindrical Near field scanning: It measures the electric field on a cylindrical surface close to the
AUT. A Fourier transform using cylindrical coordinates is used to transform the measurements to the
far-field.
• Spherical Near-Field scanning: It measures the electric field on a spherical surface close to the
AUT. A Fourier transform using spherical coordinates is used to transform these measurements to
the far-field.

The 2nd alternative to far-field testing is a compact range which is another type of far-field facility. These
are typically located indoors, using anechoic material and large reflectors. Once the radiated energy
passes the focal point of the reflector, the signal is in the far-field. Compact antenna chambers have a
“quiet zone” that defines an area in which planar waves meet the far-field criteria.

Source Antenna
antenna under test

Antenna
under test

Source
antenna

(a)
Figure 19.3. Far-field test facilities are outdoor and compact indoor (b)

Understanding the Techniques for Antenna Testing 225

Antenna Measurement System
A typical antenna range measurement system can be divided into two separate parts: the transmit site
and the receive site.
Antenna
under test
Scanner controller
Delta elevation

Sum Delta Az

B/R2 LAN

A/R2
Source 2 out
R1/R2 PNA-X receiver

Figure 19.4. Typical near field test range

When designing an antenna test facility, many parameters must be considered to select the optimum
equipment. Typically, one begins by considering the components for the transmit site, then moves to the
receive site. Designing a complete antenna system often requires you to configure the transmit site, then
the receive site, adjust the transmit site, and recalculate the values for optimum performance.

Transmit Site
The transmit site consists of the microwave transmit source, amplifiers (optional), the transmit antenna,
and the communications link to the receive site.

In selecting the transmit source, consider the frequency range of the antenna under test (AUT), the
distance to the transmit antenna, the available power of the source, and the speed requirements for the
measurements. For compact and near-field ranges, the internal VNA source will typically be the best
source to meet your measurement requirements. The internal source is faster than an external source and
may lower the cost of the complete system by eliminating a source. Large outdoor ranges may require
an external source that can be placed at a remote transmit site. Start by estimating the effective radiated
power (EIRP) of the transmitter site. The effective radiated power is the power level at the output of the
transmit antenna. To understand the required power level at the transmitter site, begin by making your
power calculations. If after doing the power calculations the transmit power is not high enough, you may
like to add an amplifier and run the calculations again.

Understanding the Techniques for Antenna Testing 226

Receive Site
The receive site consists of the AUT, a reference antenna, receiver, LO source, RF downconverter,
positioner, system software, and a computer.

The free-space loss of an antenna range determines the difference in power levels between the output
of the transmit antenna and the output of an isotropic (0dBi) antenna located at the receive site. The
test channel received power level must be calculated to determine the approximate maximum power
level present at the output of the antenna-under-test (AUT). The required measurement sensitivity is
determined from the test channel received power level, the required dynamic range, and the required
measurement accuracy.

PD = 32.45 + 20*log(R) + 20*log(F)

Where:

R = Range length (meters)

F = Test frequency (GHz)

This equation does not account for atmospheric attenuation, which can be a significant factor in certain
millimeter-wave frequency ranges.

P(AUT) = ERP – PD + G(AUT)

Where:

ERP = Effective Radiated Power (dBm)

PD = Free-space loss (dB, at the maximum test frequency)

G(AUT) = Expected maximum gain of AUT (dBi)

Measurement accuracy is affected by the measurement sensitivity of the system. The signal-to-noise
ratio will directly impact the measurement accuracy of the system for both amplitude and phase
measurements. The frequency and sensitivity requirements of your antenna system will determine the
network analyzer specifications.

Sensitivity = P(AUT) – DR – S/N – L

Where:

P(AUT) = Power at the output of the AUT (dBm)

DR = Required dynamic range (dB)

S/N = Signal-to-noise ratio (dB)

L = Cable loss (dB) from AUT to PNA input

Keysight has developed options for the PNA Series specifically for antenna measurements. Because
of these options, the PNA Series is often the preferred analyzer for antenna solutions. However, some
applications do not require these options and the lower-cost PNA-L Series or ENA Series analyzers may

Understanding the Techniques for Antenna Testing 227

be the right solution. For secure environments, a PNA or PNA-L Series analyzer must be used. Select an
analyzer from the following table that meets your frequency and sensitivity requirements.

The VNA should be located as closely as possible to AUT to minimize the RF cable lengths. The
measurement sensitivity of the PNA must be degraded by the insertion loss of the RF cable(s) to
determine system measurement sensitivity.

If the AUT is located far from the analyzer, requiring long cables, then the loss caused by the cables could
be significant, reducing accuracy and dynamic range. You may also be unable to find an analyzer that
meets your sensitivity requirements. In this situation, down-converting the signal to an IF signal by using
the LO/IF distribution unit with remote mixers brings the measurement closer to the AUT. This reduces RF
cable loss and maximizes accuracy and dynamic range.

85320A
Test mixer

85320B
Reference
mixer
LO in 85309B
IF Freq.

Amplifier

O ou
LO out

Figure 19.5. Remote mixing receive site

Measuring system speed for antennas is one of the important aspects of an antenna measuring system.
Some applications require the fastest speed a system can provide, others are concerned with the best
dynamic range available. With the PNA Series network analyzer, users can adjust their setup according to
their specific needs. The selectable bandwidth feature can be used to optimize the measurement speed
vs. sensitivity tradeoff.

Understanding the Techniques for Antenna Testing 228

By selecting the widest bandwidth available (5 MHz for DSP-4, 15 MHz for DSP-5), the measurement
speed is maximized. The PNA-X analyzer is mixer-based, with fundamental mixing to 26.5 GHz, providing
a 24 dB increase in sensitivity and dynamic range over sampler-based analyzers. This more than makes
up for the sensitivity reduction realized when the IF bandwidth of the PNA-X is opened to its maximum
to maximize measurement speed. Therefore, the PNA-X can achieve faster data acquisition speeds with
increased sensitivity in near-field applications over legacy configurations.

RCS Testing
From the radar range equation, RCS (σ) has a direct effect on the ability of a radar system to detect a
specified target at a defined range. Although the cross-section of the target cannot be controlled, the
objective in modeling RCS is to develop simulation tools capable of predicting the behavior of radar
receivers in a realistic environment.

A target’s RCS is a measure of its reflectivity in each direction, and there are three main contributors:

• Specular scattering: Localized scattering dependent on the surface material/texture and geometry
• Diffraction scattering: Incident signal scattering at target edges and discontinuities
• Multiple bounces: Reflections among target elements at offset angles

Improvements in technology have enabled a deeper understanding of how to minimize an object’s

reflected energy. As designers become more adept at minimizing σ for the smallest possible return, the
received signals are very small. The level of the returned signal is also affected by the need to use large
distances with large objects (e.g., full-sized aircraft or missiles) to ensure a planar wavefront.

Computing the IFFT on a finite-length sample produces a noteworthy artifact: It creates repetitions or
“aliases” of the fundamental signal in time. These aliases can be minimized or eliminated through a
process of testing to find an alias-free measurement span. The width of this span will depend partly on
the number of data points the analyzer can measure and process. RCS measurements tend to be very
wide frequency sweeps, ensuring the presence of band crossings. High-power pulses are often used in
RCS measurements to overcome the high losses due to low device reflection and two-way transmission
path loss. For this reason, receiver gating is often required in RCS measurements to avoid overloading the
receiver during the transmission of the pulsed-RF signal.

Source
antenna Target

Receive Target
antenna

Source/receive
antenna

(a)
Figure 19.6. RCS test facilities: outdoor far-field and compact anechoic chambers (b)

Understanding the Techniques for Antenna Testing 229

As with far-field testing, there are two main types of RCS facilities: a traditional outdoor test facility and the
compact range (as shown in Figure 19.3). RCS testing tends to be sensitive from a security perspective,
so outdoor test facilities are often in remote locations. Indoor test facilities offer optimum security but may
become large and expensive depending on the size of the target.

Simplifying Test of Phased Array Antenna

Modern EW systems and radars use phase-array antennas. A phased-array antenna is a collection of
antenna elements assembled such that the radiation pattern of each element combines with neighboring
antennas to form an effective radiation pattern called the main lobe. The antenna array is designed to
maximize the energy radiated in the main lobe while reducing the energy radiated in the side lobes to an
acceptable level. The direction of radiation can be manipulated by changing the phase of the signal fed
into each antenna element, which has an independent phase and amplitude.

The original way to implement a phased-array antenna is to use separate attenuators, phase shifters, and
other components. With the new modular approach, each antenna element and its related transmit power
amplifier (PA), receive low-noise amplifier (LNA), shifters, attenuators, and switches are packaged together
in a transmit/receive (T/R) module.

In transmit mode, the signal from the transceiver passes through the attenuator, phase shifter, and T/R
switch to the PA and then to the antenna. In receive mode, the signal from the antenna passes through
the T/R switch to the LNA and the phase shifter and attenuator to the receiver section of the transceiver.

Requirements for test and measurement must align with the application challenges we just discussed.
The testing of multiport DUTs presents four significant challenges:

• Efficiently testing all components in multiple-input/multiple-output (MIMO) antennas

• Improving measurement throughout and unit volumes while reducing cost
• Increasing the accuracy of testing for higher port-count devices
• Reducing the floor space needed for testing in component manufacturing

Various multiport measurement solutions are available today and we can divide them into two categories:
switching solutions and true multiport VNAs.

Switching solutions include a two or four-port benchtop VNA and different types of switch matrices.
Simple switching test sets are based on RF switches that route VNA ports to the various ports of the
DUT. This type of multiport solution often uses a two-port VNA to reduce cost. These test sets are
typically constructed from a combination of 1x2, 1x4, and 1x6 RF switches. Switching test sets do not
offer measurements between any two ports, so care must be taken when designing the switch tree in the
test set to ensure it meets the measurement requirements of the intended DUT.

Understanding the Techniques for Antenna Testing 230

Many multiport devices require measurements from each port to every other port. In general, the
response of any path depends upon the loading or match applied to every other port. To obtain a full
matrix of paths, a “full crossbar” switch matrix is required. In the general configuration, sets of 1xN
switch trees are cross-connected to 1x2 switches at each port. This configuration allows any path to be
measured and the unused ports are terminated with an internal load contained in the 1xN switches.

The switching-based approach has several drawbacks: complex setup, long test times, inadequate
measurement performance, and trade-offs between time and cost.

Complexity in the setup

• Switch matrices might need additional software to assist with the measurements
• Implementation becomes more difficult as the number of ports increases
• Calibration of a full N-by-N system is difficult due to the need for port terminations

Very long test times

• Because many configurations share the source and receivers, test time becomes prohibitive
• Example: 24-port switching test set supports only 144 paths, but a 24-port device has 276 paths

Given these shortcomings, it becomes vital to understand a better way to address the test challenges.
The solution is a new-generation modular VNA. It delivers three main benefits:

• Increased throughput and improved accuracy driven by enhanced speed and performance
• Enhanced configurability is well-suited to the requirements of integrated A/D subsystems
• Application versatility can satisfy the majority of the needs for the testing of A/D components

Antenna
Under
Test Locate the PXI chasis close
to the AUT to minimize
Optional RF cable losses
Trigger OUT
10 MHz

amplifier Source antenna

Trigger IN
LAN

Reference
Microwave Source Antenna
PSG/MXG/UXG
Trigger IN

10 MHz
Trigger OUT

LAN

Figure 19.7. Far field multi-channel configuration

Understanding the Techniques for Antenna Testing 231

For multi-channel antenna measurements, a multi-channel vector network analyzer is becoming quite
popular. Before the availability of multiport VNAs, an RF switch network was required to switch multiple
antenna test ports into a four-port VNA. The switch network adds additional loss (reducing measurement
sensitivity) and introduces additional measurement time to sequence through the various switch positions.
With an M980xA multiport VNA, each output port of the antenna to be measured can be routed to an
individual test receiver port, providing simultaneous measurements on every antenna port with each
trigger from the antenna rotation positioning system. This significantly reduces the measurement time for
a multiport antenna. With this multiport PXI VNA approach, there are no remote mixers utilized, so the RF
signal is routed from the antenna directly to the input ports on the PXI VNA. Because of this, minimizing
the RF cable length between the test antenna and the PXI VNA is important to minimize measurement
sensitivity degradation. Some antenna ranges are being configured with the PXI chassis located as close
as possible behind the antenna under test.

Further, to reduce the complexity of scanning a single probe antenna over a mechanical arch, multiple
probe antennas can be mounted at fixed positions on an arch, and PIN switches were used to multiplex
the different probe antennas into a four-port receiver. The AUT is then rotated slightly in azimuth and when
the AUT has been rotated 180 degrees and the entire hemispherical radiated energy from the AUT has
been measured, it’s possible to do the computation of the antenna pattern.

Figure 19.8. Measuring antenna patterns with multiport receivers

Understanding the Techniques for Antenna Testing 232

This method reduces the complexity of the mechanical positioning required for a single probe antenna but
adds complexity and loss with the RF switches.

Keysight Technologies provides many of the components required to make accurate antenna and radar
cross-section (RCS) measurements. Keysight instruments provide the greatest accuracy, reliability, and
productivity available.

You can integrate our instruments into your antenna test systems to measure a wide range of data
acquisition speeds and measurement sensitivities. Applications include near field, far field, and radar
cross section measurements.

Understanding the Techniques for Antenna Testing 233

Section 5

System Level Radar and EW Test

Introduction
Today, different types of radar systems are used in a variety of applications: avionic, military,
automotive, law enforcement, astronomy, mapping, weather, and more. Within this broad range of
uses, several radar technologies have emerged to meet specific needs in terms of performance,
cost, size, and capability. For example, many police radars use continuous-wave (CW) radar to
simply assess Doppler shifts from moving cars; range information is not needed. As a result, low
cost and small size are more important than advanced capabilities and features.

In this section, learn how to simulate an effective and accurate radar environment, and understand
how these processes are innovative in the radar and EW industry. Various system level test and
measurement procedures are covered including validation of jammers effectiveness, Electronic
Intelligence, signal recording and analysis. The section highlights the critical need of system
calibration, field testing, sustainment and services. It also touches upon the importance of software
testing and new age cyber security.
Chapter 20
Avoiding 5G Coexistence and Effectively Simulating
Background Signals
With new bands specified for 5G communications systems, there is now an issue of coexistence,
which is radar and satellite using the same or similar frequency bands with 5G. Such coexistence
can cause loss of capacity in 5G systems and can even damage sensitive front ends in satellite
ground stations.
There are three primary use cases envisioned for 5G communication systems.

ENHANCED MOBILE
3D Videos, UHD
BROADBAND

20 Gbps
Augmented reality
Smart bldg.

MASSIVE MACHINE
5G ULTRA-RELIABLE
Industry automation

TYPE COMMUNICATION LOW LATENCY COMMUNICATION

10M devices/km2 1 ms Latency Telemedicine

10 year battery life 99.9999% Reliability

Self-driving car
Smart city
Mission-critical

Figure 20.1. 5G use cases

The topmost hexagon is the familiar use case of watching high-definition videos on your phone. This
could entail 3D videos or augmented reality. The lower right hexagon use case is ultra-reliable low latency
communications. This would entail 1ms latency, roughly a 10x improvement, and 99.9999% reliability.
Some of the applications include industrial automation, telesurgery, self-driving vehicles, and mission-
critical communications. The lower left hexagon is massive machine-type communication. This is akin
to IoT sensors, where you have 10 million sensors per square kilometer. The data that they pass back is
meant to be low rate, therefore low bandwidth, and less power consuming. The battery in these devices
is meant to last for 10 years, and after this time, the sensors are thrown away.

These 3 use cases signify different frequency bands. The enhanced mobile broadband service needs
higher bandwidth, and higher bandwidth means higher frequency which tends to be 20+ GHz or higher.
The ultra-reliable low latency communications require mid-band frequencies, roughly 3 to 6 GHz. The
massive machine-type communications need low bandwidth to conserve power, so the frequency band
tends to be < 3 GHz.

mMTC UR/LLC eMBB

106 devices/Km2 1ms latency 5 to 20Gbps

3GHz 6GHz 100GHz

Low band Mid band High band

Figure 20.2. 5G NR spectrum targets for different use cases

Avoiding 5G Coexistence and Effectively Simulating Background Signals 237

5G operating bands are described by Frequency Range 1 and Frequency Range 2 (FR1 and FR2).
Frequency Range 1 is also called sub-6 GHz, and incorporates the low-band and mid-band. FR2 includes
bands around 28 GHz and 39 GHz and is also called mmWave. More bands will be allocated as 5G gets
further revisions to the standard.

Low Bands < 1 GHz Mid Bands 1 GHz - 6 GHz High Bands > w24 GHz
(Freq Uplink, Freq Downlink (Freq Uplink, Freq Downlink (Freq Uplink, Freq Downlink
unless otherwise noted) unless otherwise noted) unless otherwise noted)

5G NR n5: 824 - 849 MHz, 869 - 894 MHz n1: 1920 - 1980 MHz, 2110 - 2170 MHz n66: 1710 - 1780 MHz, 2110 - 2200 MHz n257: 26500 - 29500 MHz,
n8: 880 - 915 MHz, 925 - 960 MHz n2: 1850 - 1910 MHz, 1930 - 1990 MHz n70: 1695 - 1710 MHz, 1995 - 2020 MHz 26500 - 29500 MHz
n12: 699 - 716 MHz, 729 - 746 MHz n3: 1710 - 1785 MHz, 1805 - 1880 MHz n74: 1427 - 1470 MHz, 1475 - 1518 MHz n258: 24250 - 27500 MHz,
n20: 832 - 862 MHz, 791 - 821 MHz n7: 2500 - 2570 MHz, 2620 - 2690 MHz n75: 1432 - 1517 MHz (DL) 24250 - 27500 MHz
n28: 703 - 748 MHz, 758 - 803 MHz n25: 1850 - 1915 MHz, 1930 - 1995 MHz n76: 1427 - 1432 MHz (DL) n260: 37000 - 40000 MHz,
n71: 663 - 698 MHz, 617 - 652 MHz n34: 2010 - 2025 MHz, 2010 - 2025 MHz n77: 3300 - 4200 MHz, 3300 - 4200 MHz 37000 - 40000 MHz
n81: 880 - 915 MHz (SUL)
n38: 2570 - 2620 MHz, 2570 - 2620 MHz n78: 3300 - 3800 MHz, 3300 - 3800 MHz n261: 27500 - 28350 MHz,
n82: 832 - 862 MHZ (SUL)
n39: 1880 - 1920 MHz, 1880 - 1920 MHz n79: 4400 - 5000 MHz, 4400 - 5000 MHz 27500 - 28350 MHz
n83: 703 - 748 MHz (SUL)
n40: 2300 - 2400 MHz, 2300 - 2400 MHz n80: 1710 - 1785 MHz (SUL)
n41: 2496 - 2690 MHz, 2496 - 2690 MHz n84: 1920 - 1980 MHz (SUL)
n50: 1432 - 1517 MHz, 1432 - 1517 MHz n86: 1710 - 1780 MHz (SUL)
n51: 1427 - 1432 MHz, 1427 - 1437 MHz

Figure 20.3. Sub-6GHz and mmWave operating bands in 5G NR

There are many new operating bands in 5G. In particular, the n77 band spans 3.3 to 4.2 GHz and n78
band spans 3.3 to 3.8 GHz. Many new 5G commercial deployments are focused on these bands. This
overlaps with a possible downlink frequency range of satellite ground stations, from 3.4 to 4.2 GHz.

I
ISM ISM 64-71GHz
0.6 GHz S 2.5 GHz 3.3-4.2 GHz 4.4-5 GHz 5.7-5.8 24 GHz
24 GHz 28 GHz 39 GHz
M 71-76GHz

Frequency Range 1: 400 MHz to 6 GHz Frequency Range 2: 24.25 to 52.6 GHz

Adds 1.5 GHz of new spectrum in Adds 8.5 GHz of new spectrum in Frequency up to 90 GHz
frequency bands frequency bands are currently being
investigated
for future releases.
n77: 3.3-4.2 GHz n257: 26.5-29.5 GHz
n78: 3.3-3.8 GHz n258: 24.25-27.5 GHz
n79: 4.4-5 GHz n260: 37-40 GHz

Figure 20.4. New operating bands in 5G NR

Avoiding 5G Coexistence and Effectively Simulating Background Signals 238

In addition, the n257 band spans 26.5 to 29.5 GHz, and the n258 band spans 24.25 to 27.5 GHz.
The n260 band spans 37-40 GHz and the n261 band spans 27.5 to 28.35 GHz. These are all
overlapping with fixed satellite services ground station uplinks at the 27.5 to 29.5 GHz and downlink at
37.5 to 40 GHz.

4.4-4.9 64-71 GHz

0.6 GHz 2.5 GHz 3.3-3.8 GHz ISM 28 GHz 39 GHz
GHz 71-76 GHz

3.5GHz IMT
5G

3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 4.0 4.1 4.2 GHz

Radio location Extended C-Band Standard C-Band

services (e.g. radar)
Fixed Satellite Service

Region 1 3.5 GHz IMT around the world

3.3-3.4 GHz Africa, some countries in regions 2/3

3.4-3.6 GHz Regions 1,2 and most of 3

Region 2 3.6-3.7 GHz Some in region 2 and 3

3.6-3.8 GHz EU
Region 3

Figure 20.5. Detailed frequency range for 5G NR

Currently, in the US, the entire 3.1-3.55 GHz band is allocated for both federal and non-federal
radiolocation services. Federal radiolocation services have a primary allocation and non-federal users
operate on a secondary basis. In addition, the C-band and extended C-band are used for fixed satellite
services. This introduces a conflict condition between 5G and radar/satellite allocations which is known
as coexistence.

Radar Coexistence with 5G

Coexistence is similar to interference, except that two or more signals have the right to occupy the
spectrum in question. However, usually one of the signals has priority, typically radar has priority over
5G. Thus, 5G must either shut off or move off frequencies to account for this. The coexistence of 5G
and satellite is dangerous because satellite ground stations have very sensitive RF front ends that are
designed to receive very lower-power signals from an altitude of 36,000 km. This makes receiver front
ends very sensitive to external interference. On the other hand, 5G base stations are strong as they are
designed to speak to phones from one to two miles away.

Avoiding 5G Coexistence and Effectively Simulating Background Signals 239

Due to coexistence, it is incredibly important that there be a set of concrete metrics that can be used
to assess the signal quality and thus the impact of coexistence. One of the key parameters for signals,
in general, is error vector magnitude. Error vector magnitude is the measure of the difference between
a measured symbol and the reference symbol in two dimensions (I and Q). The higher the EVM is, the
poorer the demodulation will be. A perfectly clean signal will theoretically have 0 EVM. EVM is often
expressed in dB or percentage.

IQ Error
Q Magnitude Vector
Error

IQ measured
Q err EVM [n]
Ø Error Vector
Phase

IQ reference
I
IQ Phase I err
Error

Figure 20.6. Error vector magnitude (EVM)

The 3GPP standard for 5G has laid out EVM requirements for different modulation schemes because the
modulation can be changed depending on what the channel can support. If the channel is cleaner and
can support a higher-order modulation, which is more signals being transmitted, then it will. In Table 20.1,
you can see that the QPSK, the least order modulation, has the highest required EVM, followed by 16
QAM, 64 QAM, and then 256 QAM, which requires a clean channel.

Modulation scheme for PDSCH Required EVM

QPSK 17.5 % (-15.1 dB)
16QAM 12.5 % (-18.1 dB)
64QAM 8 % (-21.9 dB)
256QAM 3.5 % (-29.1 dB)

Table 20.1. Table shows how 3GPP EVM requirements for user equipment (UE) decrease as the modulation density
increases

Avoiding 5G Coexistence and Effectively Simulating Background Signals 240

There are several discriminators for minimizing coexistence conditions. The frequency regulator could
set up guard bands and distinct frequency spacing of different services to prevent coexistence. Services
could dictate a minimum distance from transmitters. For example, shipborne radars could say that they
need to be X miles away from the coast where there may be 5G base stations. Also, the power level can
be adjusted to minimize coexistence. Lastly, antenna type, angle, and elevation is another parameter to
adjust. Of course, the measurements that would assess coexistence impact are EVM and throughout
testing or satellite video signal quality.

A radio channel is the propagation path between a transmitter and a receiver. It includes such effects as
path loss, shadowing, multipath and fading, and Doppler.

PROPAGATION
Mobile station

Base station

Total Signal
Path Loss

Mobile route

Shadowing

MS

Fast Fading BS

f
Doppler spread

Figure 20.7. Radio channel diagram

The superposition of multi-path components causes fast fading. This is when phases of multi-paths are
independently varying and summing together. Distance-dependent attenuation of propagating signals is
called path loss. The longer the propagation path, the greater the path loss. Large obstacles between the
transmitter (Tx) and the receiver (Rx) create shadowing and cause slow fading. A fading channel affects
RF signals between UE and BTS (over-the-air).

As the threat emulation environment is built up, one of the first considerations needs to be the time-base
of the system; specifically all emitters need to be phase-locked to the same time base. A fully coherent
system is critical for a successful model of threat emulation. In summary, coexistence is between 5G
and radar/satellite and is a very real problem in a crowded spectrum. The impact of 5G on radar/satellite
Avoiding 5G Coexistence and Effectively Simulating Background Signals 241
and vice versa can be measured by EVM or by 5G quality metrics. An adequate guard band should be
chosen to minimize the issue to the extent possible. Power levels, distance, and antenna orientation are
other mitigation parameters. With the ability to measure coexistence, the effects of it may be lessened to
provide seamless communication services.

Simulating Non-Radar Background Signals

An important task for an EW receiver is to detect incoming signals (waveforms), identify exactly which
station or stations they are coming from, and analyze signal information from all transmission stations. The
information coming from a signal transmission station includes the station’s location, speed, waveform
types, and frequency bands. The signal appearing at the EW receiver input is a combination of signals
from different radar or communication transmission stations with complex information for the location and
speed of the stations, as well as time waveforms and the frequency bands of transmitted signals. To test
the EW receiver, a test signal with the following characteristic is needed:

• It must come from multiple radar and communication transmission stations

• Each component of the EW receiving signal must include information from the transmission station
on its location and speed, as well as the time waveform from the station and the signal’s frequency-
domain information
• It must form multi-emitter, overlapping, or non-overlapping signals

This type of test signal is called a Multi-Dimensional (MD) signal.

Figure 20.8. Shown here is an example of an ew receiver test environment with several geographically dispersed
emitters contributing to signal conditions at a centrally-located receiver.

Avoiding 5G Coexistence and Effectively Simulating Background Signals 242

Testing an EW Receiver
To test an EW receiver, a test signal must be generated and that does not mean simply adding several
time waveforms together. Instead, a MD signal must be created using a detailed setup. For example,
because the EW receiver might be installed on an airplane, car, or ship, the tool that’s used to generate
that test signal must allow the user to specify the EW Rx station’s location, speed, time waveform, and
frequency band. Also, for each radar station, the tool must allow the user to specify its location, speed,
time, waveform, and frequency band, as all this information is built into the MD signal.

There are 4 steps to generate the test signal:

Step 1: Generate the Tx signal for each Tx with location described by longitude, latitude, and height, as
well as speed, the proper time waveform, and frequency content (carrier and Doppler frequency). The
required complex Tx comprises:

• An antenna and active array antenna with beamforming

• Pulse and dynamic pulse
• Environment scenarios
• A multi-emitter from radar and communication systems
• An EW receiver test signal with MD information from the radar stations
• A long test sequence with a wide frequency band

Step 2. Determine EW receiver location. Speed also needs to be considered.

Step 3. Combine all Tx signals together to form the MD signal.

Step 4. Prior to the test, verify the MD signal.

EW Test Platform
To test the EW receiver, a test platform is needed in which the MD signal can be built and analyzed. The
proposed system will work under actual tactical situations with terrain, numerous threats and targets, plus
multiple radar signals and jammers. SystemVue can provide a design and test platform with radar/EW
scenarios such as target radar cross section (RCS), interference, jamming, deception, and clutter. With
SystemVue, an AGI System Tool Kit (STK) link is provided. This link allows the user to describe complex
radar/EW environments under actual tactical situations with terrain, numerous threats and targets, plus
multiple radar signals, and jammers simultaneously occurring.

One of the challenges that is often encountered when testing radar/EW systems is how to generate and
analyze the MD signals, which include information about monitored Tx stations, such as location, velocity,
time, and frequency. Radar and EW environments also include interference, jamming/ deception and this
must be taken into consideration when performing measurements under real-world scenarios. A solution
to address these issues using SystemVue has been proposed, and simulation and test platforms have
been built. Using these platforms, engineers gain access to a myriad of benefits. They provide a true
design-oriented value proposition to shorten the development cycle and allow users to save time and
money by minimizing field tests. Moreover, SystemVue’s multiple environment scenarios enable engineers
to create real-world test environments that enable them to design high-quality products. Such capabilities
and benefits are critical to ensuring the successful development of modern radar and EW systems.
Simulation and test results show custom problems can be solved using the proposed method.

OVERVIEW OF KEYSIGHT 5G TESTBED

https://www.keysight.com/see/radar-ew-system-test

Avoiding 5G Coexistence and Effectively Simulating Background Signals 244

Chapter 21
Validating Jammer Effectiveness
Test systems used to evaluate jammer effectiveness are structured so that each component,
depicted as a block in the diagram below, has a specific function and relation with the System
Under Test (SUT).

Radar Emulation
Scenario Controller
Channel Emulation - PDW Domain Radar Echo Generator

Tx Emulation

Rx Emulation
Channel Emulation - PDW Domain Signal Generator

Radar Rx Front End

Signal Accumulator (AGC) Emulation
Wideband
Digitizer
Sysytem Under Test
Channel Emulation - Time Domain Signal Conditioning

Figure 21.1. A candidate test system to evaluate jammer effectiveness

Validating Jammer Effectiveness 245

The overall operation of the system shown in Figure 21.1 is described in the following paragraphs.
This ‘thought experiment’ will highlight the types of challenges that will be faced by the implementers of
such a system.

Scenario Controller
A scenario controller is a software application (often supplemented by a GPU) that keeps track of the
actions of all the objects in the emulated scenario – their positions, velocities, aspects, and relationships
with one another. Such applications are broadly available although many of them are not architected for
closed-loop operation and can only create a set of signals that would be applied to the SUT, without the
ability to emulate a radar in real-time.

It is here that we first encounter challenges in the implementation of this test system:

• The SUT may function and maneuver autonomously, without the knowledge of the scenario controller.
This presents the need for a two-way communication path between the SUT and the scenario
controller. The scenario controller needs to know where the SUT is at any given time, and the
SUT needs to know about the surrounding terrain and other emulated objects so that it can
properly maneuver.
• Timing is key – precise timing between all system elements is critical, including the SUT.

Radar Transmitter
Radar emulation is central to the operation of this test system. Depending on the purpose for which the
system is to be used, radar emulation may be high-fidelity (emulating a specific radar model) or more
general-purpose in nature. In any event, the radar emulator needs to be reconfigurable so that it may be
used to emulate more than one kind of radar.

In the system architecture used here, the radar transmitter emulator’s output is a series of pulse
description words (PDWs), which describe the pulses that the radar wishes to transmit. The PDWs will
change according to the mode of the radar, and the radar emulator must transmit certain information
(such as antenna beam direction) to other test system elements.

Radar emulation will likely be implemented with a combination of software and FPGA-based firmware,
requiring the ability to reconfigure FPGA contents at will. This will require a new technology development.

Channel Emulation (PDW Domain)

The PDW-based output of the radar transmitter emulator must be scaled to account for the signal path
between the radar and the SUT. For simple cases, this may be a simple delay and attenuation factor.
More advanced systems will include models for multipath transmission, general scattering, atmospheric
absorption, and other factors that may affect the transmitted signal. Channel models must be applied to
the PDWs to account for these effects.

Validating Jammer Effectiveness 246

But the SUT must participate in channel emulation, and there are other logistical problems:

• The test system knows the beam direction of the radar but not the beam direction of the SUT. This
requires a communication path between the SUT and the transmit channel emulator to properly scale
the SUT’s input signal.
• The SUT antenna might be an array. Full, element-by-element testing of phased array antennas is
possible but expensive. Alternatively, the signal may be applied to the SUT at a point behind the
antenna if a physical connection is available.
• The SUT might be trying to execute a direction-finding algorithm, requiring a test setup specifically
designed to enable direction-finding (with a calibration routine), or the ability for the test system to
‘push’ direction information into the SUT’s controller via a special test mode.

Signal Generator
Fortunately, several suitable signal sources have become available in recent years. Such sources accept
the PDW output of the channel emulator and create RF signals to match, which can be applied to the
input of the SUT.

Signal Conditioning
The SUT’s output must be assumed to be unpredictable and of varying power levels over time. It
is important to prevent the SUT’s power, which may be quite high, from damaging the input of the
downstream receiver circuits. At the same time, the receiver’s input power should be in a range that
allows the digitizer to operate near full scale so that optimum signal fidelity is maintained.

Signal conditioning can be achieved using a fast gain control circuit (with power protection at the receiver
input to handle any transient signals). Preferably, a software interface with the SUT would allow the test
system to determine appropriate attenuation levels – obviously requiring cooperation from the SUT.

There is an additional, and unusual, requirement on the signal condition solution needed for this test
system. To enable proper signal scaling downstream, the signal conditioning system must communicate
its attenuation setting to the rest of the test system. The signal conditioning system must therefore include
the appropriate communications mechanism while also participating in system timing functions.

Digitizer (Receiver)
Assuming successful implementation of signal conditioning at the SUT’s output, the receiver design
may be straightforward. The block diagram in Figure 21.1 assumes the use of a wide-band digitizer of
a type that has become recently available off-the-shelf. This type of design does not require a tunable
downconverter and may make it easier for the system to handle frequency hopping SUTs. A more
traditional downconverter-based design (with a narrowband digitizer) could also be used, in which case
the test system must be able to follow frequency changes by the SUT. In either case, an interface with the
SUT that allows the test system to follow frequency hops more easily is desirable.

Validating Jammer Effectiveness 247

Channel Emulation (Time Domain)
The SUT’s captured output signal must be modified to account for channel effects over the path from
the SUT to the radar. Although conceptually identical to the channel emulation already discussed, the
implementation here is different because the input signal consists of time-domain sampled data (probably
baseband IQ data), not PDWs. This requires considerably more processing power.

Commercial channel emulators are available which can perform this function, but their inputs and outputs
are usually RF signals, and they are designed for mobile phone applications, so they are not useful in the
context of the test system examined in this paper. The underlying technology is available but must be re-
implemented in the proper form.

Radar Echo Generator

The total set of signals in the simulated scenario must include radar echo returns. In the very simple
case illustrated in Figure 21.1, there are two echo returns: one for the jammer and one for the protected
entity. These echo returns must be mathematically generated in software, with channel models applied.
These signals are represented as PDWs to keep data bandwidths low, and the channel models are like
the functionality already described for the radar’s transmitted signals except for the addition of round-trip
effects rather than just one-way. The SUT is not involved in this step.

Signal Accumulator
The captured signal from the SUT must be combined with the emulated radar echoes, using precise
timing. The signal accumulator block represents a new (but straightforward) technology that must be
developed. Note that the echo signals are in the PDW domain while the SUT’s output is captured in the
time domain, so the signal accumulator must translate PDWs into the time domain. In addition, the SUTs
signal will not arrive at the same time as the emulated radar echo signals, so buffering and careful timing
is necessary. Once accomplished, the output of the signal accumulator is a time-domain signal that
represents the RF signal that would be captured at the radar’s antenna.

Radar Receiver Emulation

It is important to properly emulate the radar receiver, including automatic gain control at the receiver’s
inputs. Many jamming techniques rely on the jammer’s ability to manipulate the radar’s AGC so that
actual echoes are driven down into the noise floor while the jammer’s signal is stronger. Radar receiver
algorithms must also be implemented. However, the necessary fidelity of radar emulation depends on
the specific use case – some systems will require detailed, high-fidelity emulation of specific radars, while
others only need general-purpose radar signals of different types and do not need to exactly match any
particular model of radar.

Validating Jammer Effectiveness 248

Jammer Effectiveness Evaluation
A survey of the industry indicates that there is no standard method for evaluating the effectiveness
of a jammer. Jammer effectiveness is at its core a mission-specific measurement. If a jammer can “fool”
a radar for five minutes, is that enough? It’s easy to see that the answer depends on the specifics of
the mission.

More generally, jammer effectiveness should be evaluated on a statistical basis – especially since the
jammer may ‘learn’ from one simulated run to the next, and its behavior will be modified by noise,
different navigation paths, and its cognitive algorithms.

A standard measurement technique would be helpful. Such a standard would help to avoid confusion and
duplication of effort across the field. For now, no such standard exists. This should be the topic of future
papers and industry discussions.

System Scalability
It is important to consider the issues that will arise when a test system is scaled up to include multiple
radars, jammers, and protected entities. Noting that all radar signals must be applied to all jammers, all
jammer signals must be applied to all radars, and all radar echoes must be received by all radars, it can
be seen that the complexity of signal and data connections in the system will expand factorially as the
system is scaled up. This presents architectural problems that must be resolved to prevent the system
from becoming unmanageable

As a subject for future investigation, it is proposed that ring and star architectures be implemented.
In these types of architectures, all of the necessary data is broadcast to every part of the test system.
System blocks can then listen for the data they need, ignoring the rest. Such architectures are often
used to create scalable systems – but these architectures are not common in test and measurement
applications, and further investigation will be needed.

Testing cognitive jammers – or any other cognitive system – will require the same level of innovative
thought that’s required to design the cognitive systems in the first place.

Cognitive systems will have to be full participants in their testing, with software and hardware interfaces
included that allow the test system to communicate with the SUT in a peer-to-peer fashion. Designers of
cognitive systems will be well-served to consider these requirements early in the design and development
process. Otherwise, expensive modifications will be needed to test them later.

Validating Jammer Effectiveness 249

Likewise, test system designers will be forced to implement new technologies and new architectures –
and the costs of such systems must be included in budgets at an early stage.

ELECTRONIC WARFARE SIMULATION SOLUTION

COURTESY OF MICROWAVE JOURNAL
https://www.keysight.com/see/radar-ew-system-test

Validating Jammer Effectiveness 250

Chapter 22
Radar and EW Signal Recording and Analysis
Gap-free recording of a complex RF environment is a practical way to capture elusive and
intermittent signals for further analysis via playback and post-processing. The Keysight N9040B
UXA and N9030B PXA signal analyzers enable streaming of I/Q data at up to 255 MHz real-time to
the X-COM IQC5255B signal record and playback system (Figure 22.1).

X-COM Systems
IQC5255B

LAN PCIe®
LVDS

Keysight UXA Signal Analyzer

N9040B

Figure 22.1. Combining the UXA signal analyzer and IQC5255B recorder with powerful software components
creates a comprehensive solution for RF streaming.

Radar and EW Signal Recording and Analysis 251

With up to 15 TB of memory, the IQC5255B can capture more than three hours of data at full bandwidth
Table 22.1 provides approximate recording times based on the bandwidth setting in the UXA.

Approximate Recording Time

Bandwidth Memory Seconds Hours Days
10 MHz 15 TB 300 k 83.3 3.5
100 MHz 15 TB 30 k 8.3 0.35
255 MHz 15 TB 12.5 k 3.5 0.14

Table 22.1. Maximum recording time is a function of available memory and analyzer bandwidth setting.

Configuring and Recording Seamlessly

When the environment contains numerous emitters of varying amplitudes, it can be difficult to find low-
level signals in the presence of much larger ones. In the streaming solution, the PXA or UXA provides 78
dBc of spurious free dynamic range (SFDR) across the full 50 GHz frequency range, and this is recorded
with the high-resolution 16-bit I and Q capture capability of the X-COM recorder to produce deeper views
of signal behavior.

Seamless integration between the analyzer and recorder enables you to configure and initiate recordings
directly from the UXA or PXA, streamlining recording control and eliminating the need for an external PC
or laptop. An added benefit is the ability to simultaneously view the live spectrum measurements on the
analyzer screen while a recording is in progress (Figure 22.2).

Figure 22.2. Real-time spectrum analysis (RTSA) mode provides live measurements of elusive signals even while
recording is underway.

Radar and EW Signal Recording and Analysis 252

Extracting Useful Information from Massive Data Sets
This is a “big data” process. Streaming 100 MHz of I/Q bandwidth with 16-bit samples for just 10 minutes
creates a 300 GB file. Today’s RF streaming applications routinely require bandwidths above 200MHz.

With a modern interface such as PCIe, large files can be offloaded from the IQC5255B’s redundant array
of independent drives (RAID) array to the analysis computer (in the UXA) with nearly gigabyte-per-second
speeds. Once the data has been captured and stored as I/Q samples on the IQC recorder, it must be
analyzed, interpreted, and distilled into more compact forms such as mission data file (MDF) entries or
pulse descriptor words (PDWs).

Advanced signal processing tools are needed to turn this raw data into useful information and actionable
results. For example, the ability to search through large sets of data and isolate specific signal behavior
is essential to helping a system engineer or signal analyst determine whether the system performed as
expected or if there were any timing anomalies or unexpected RF emitters.

X-COM’s Spectro-X software allows you to quickly zoom in on desired recording segments using
automated signal- and pulse-search algorithms. It includes functions that range from basic frequency,
time, or amplitude filtering to highly advanced pulse pruning based on pulse width, pulse repetition
interval (PRI), and more. This level of functionality makes it easy to search through vast sets of data and
uncover rare events or signals (Figure 22.3).

Figure 22.3. Whether viewed in the time (top) or frequency domain (bottom), Spectro-X software helps detect
short-lived signal behaviors as in this interaction between an interceptor radar and a jamming response. Note the
relatively short duration of the interaction compared to the other scanning radars in the background.

Radar and EW Signal Recording and Analysis 253

With the real-time streaming option on the UXA and PXA signal analyzers (Option RTS) combined with
the X-COM Systems IQC5255B, you can now capture and record I/Q signatures of complex pulsed radar
signal environments, enabling signal characterization and system verification. This comprehensive solution
also includes scenario emulation through the ability to play recordings through a vector signal generator.
With seamless integration and accessible usability, the streaming solution from Keysight and X-COM can
help you push the envelope in next-generation radar and electronic warfare systems.

RF Streaming for Aerospace & Defense Applications

Some systems record at either RF or IF, necessitating digital down conversion to baseband. Whether the
down conversion occurs digitally or using RF circuitry, Figure 22.4 below highlights the process for turning
large volumes of raw IQ data samples to mission data file entries or pulse descriptor words.

Electromagnetic Environment
Raw
IQ Data

Record 100% of Signal Content

(minutes, hours, days)
Recording Hardware extended periods of time

Identify Signals
DSP Tools of Interest Operate on very large data sets

Replay
Re-generation Hardware Portions
Recreate with high signal purity

Info Analyze/ Leverage existing analysis tools

DSP Tools Demod using open-format data files
Figure 22.4. RF energy is first stored as IQ samples where DSP tools are used to identify and demodulate signals
of interest. These stored IQ samples can also be used for signal playback or combined with other signal files for
scenario simulation.

RF streaming is big data. With 16 bit resolution on I and Q, streaming 100 MHz of IQ bandwidth for just
10 minutes would create 300 GB of data. Modern RF streaming applications routinely require bandwidths
over 200 MHz.

Thanks to modern digital interfaces and fast solid-state disk drive technology, recordings can be
streamed directly to the disk from the digitizing hardware. Wider streaming bandwidths require the data
stream to be split and fanned out to slow down the individual write rates to each disk. The parallelization
of data writing (and reading) is a storage management technique for a RAID.

Radar and EW Signal Recording and Analysis 254

 RF Conversion I/O Data Routing Storage
Frequency Data Data Storage RAID
System Control
Translate Convert Management Management Arrays
RAID
RF

and Routing
ADC

Data Sync
Real/IQ Controllers
In FPGA
(Processing /
Filtering)
RF DAC
RAID
Real/IQ Controllers
Out
High-Speed Offload

Analytics Workstation

Figure 22.5. Each link in the streaming solution is designed to meet or exceed data throughput requirements

RF Streaming Collection Model

Figure 22.6 below is an example of a recording environment where all the above issues will come into
play. In this case, the RF streaming solution will be on-board the Northrop Grumman E2-C Hawkeye
surveillance aircraft, with a mission of recording the RF interaction between the interceptor aircraft radar
(blue) and the UAV jammer (red). The data collected will also include the two surveillance radar signatures
(blue) occurring in the background.

UAV Jammer
mer

Surveillance
AIRCRAFT
ing
eam Base
Str SURVEILLANCE
S
RF

R
h RADA
rray Searc Interceptor
nte
Base Pha sed A DA
R AIRCRAFT
RA
SURVEILLANCE ch-
ear
or S
ect
Bis

Figure 22.6. An RF streaming solution is placed on-board an E2-C to record the interaction between interceptor
radar and jamming response from the UAV.

Radar and EW Signal Recording and Analysis 255

RF Streaming Requirements
There are four main performance considerations in any RF recording system

• RF frequency coverage and analysis bandwidth

• Spurious Free Dynamic Range
• Record time
• Metadata inclusion

The first is the amount of bandwidth needed to capture the signals of interest. Even for multi-channel
recording systems, the engineer must determine how much IQ bandwidth is needed to capture the
signal in its entirety. Extracting pulse descriptor words from a signal that straddles more than channels is
very difficult. Furthermore, the streaming bandwidth should exhibit flat amplitude response and excellent
phase linearity to minimize distortion to the data. Poor IF flatness will manifest as poor ACPR or EVM on a
communications signal or phase errors on chirped radar pulses.

The second determinant in recorder selection is SFDR, which represents the lowest amplitude signal
that may be discerned from a large interfering signal. As defined by the ratio between the power of the
carrier signal and the RMS power of the next most significant spurious signal, it represents the sensitivity
of the measurement system to small signals. In pulse-Doppler radar signal processing, targets are often
represented in terms of their range and velocity.

Radar and EW Signal Recording and Analysis 256

Simulating the Threat Environment
RF streaming on mission day should be, well, boring. To complete the RF streaming portion of the
exercise, the operator needs to collect the required IQ signatures and bring these back so the system
engineer or signal analyst can begin data analysis. In this regard, it is important to prepare for what you
expect to happen, but also to plan for what you don’t. Figure 22.7 below illustrates our expected radar-
jammer interaction on mission day. Note the timing relationship between the Blueforce (blue pulses) and
Redforce (red pulses) signatures. Two important timing measurements are as follows.

• The response time between the first Blueforce transmit pulse and the first reply by the
Redforce jammer.
• The reaction time of the Redforce EW system, between when the Redforce radar warning receiver
(RWR) recognizes a potential threat and its first reply.

A live measurement cannot be performed on mission day, since this would require access to the Redforce
EW system test points. However, this measurement can be performed in a chamber if both Redforce and
Blueforce systems are available.

Figure 22.7. Radar-Jammer response timing. Blueforce initiates pulses which are detected by the Redforce RWR.
The Redforce then initiates a credible jamming signal as quickly as possible. Both Redforce reaction times are
critical measurements.

Radar and EW Signal Recording and Analysis 257

These emitters and their relative timings may be simulated by combining the RF outputs of two precision
signal generators and feeding their combined output to the receiver of the RF streaming system. This is
shown below in Figure 22.8 with the environment simulation on the left and the RF streaming solution on
the right. The initial Blueforce and Redforce pulses can be time tagged by connecting the trigger outputs
from the RF signal generators to the marker inputs of the IQC5255B recorder streaming solution.

Figure 22.8. Example setup to simulate the Blueforce (radar) and Redforce (Jammer) timing and event recording.
The IQC5255B recorder accepts both marker and trigger inputs from the signal generators, along with high-speed
LVDS IQ data from the UXA signal analyzer.

Post-Mission Analysis and Simulation

After data has been collected, questions abound. What happened on mission day? Figure 22.9 below
shows the magnitude response of a 3-second interaction between the interceptor radar and the jamming
response. Note the relatively short duration of the interaction compared to the other scanning radars in
the background.
Blueforce Radar and
Redforce Jammer Interaction
2020

1010

00 Other Surveillance Radars

Magnitude (dBm)

-10-10

-20-20
(d B m )

-30-30
Ma g n itu d e

-40-40

-50-50

-60-60

-70-70

-80-80
0.0 0.3 0.6 0.9 1.2 1.5 1.8 2.1 2.4 2.7 3.0
0.0 0.3 0.6 0.9 1.2 1.5(sec )
T ime 1.8 2.1 2.4 2.7 3.0
Time (sec)

Figure 22.9. Magnitude profile of RF environment during radar-jammer interaction. The top yellow trace is the
maximum value of the magnitude response while the lower green trace is the RMS value.

Radar and EW Signal Recording and Analysis 258

By zooming into the time-domain profile of the Blueforce/Redforce interaction, it is easy to see the relative
timing between the first Blueforce pulse and that of the Redforce (Figure 22.10). Precise measurement
of the timing relationship between the two emitters can be made with plot markers. Also, note the signal
anomaly that appears after the fifth Blueforce pulse. The question for the system engineer to determine is
whether the anomaly is radiating from the Blueforce aircraft or the Redforce.

Figure 22.10. Successive zooming on the time-domain plots can reveal signal details down to the sample level.

Once a time segment has been identified for study as in Figure 22.10 above, the interaction can be
replayed in both frequency and spectrogram (joint time-frequency) domains for visual inspection.

Figure 22.11. Persistence density plot of emitter jammer interaction.

Radar and EW Signal Recording and Analysis 259

Areas of interest may be further analyzed by Keysight’s 89600 VSA (vector signal analysis) software. A
summary of the pulse signals contained in the zoomed time range above is shown below in Figure 22.12.

Figure 22.12. Automatic pulse parameter calculation speeds validation of PDW performance against design
objectives.

Using the spectrogram function in the signal processing toolbox, we can reproduce the time-frequency
plot shown above, but in 3-D. Below, we indicated the signal anomaly with a marker. This is likely a
numerical artifact, as we find it shows up at the frequency limit of –100 MHz.

Figure 22.13. Spectrogram plot of emitter-jammer interaction.

Radar and EW Signal Recording and Analysis 260

RF streaming provides system engineers undeniable proof of what happened over long time durations.
Keysight and X-COM Systems have teamed to develop a wideband, integrated solution that offers
exceptional fidelity (SFDR and IF flatness), IQ data resolution, analysis bandwidth, RF frequency range,
and gap-free recording capacity. While other vendors offer all-in-one packages as well, the X-Series signal
analyzers and IQC5000B recorders offer the highest recording performance in the industry. Furthermore,
with software tools such as Spectro-X, 89600 VSA software, and MATLAB, users can accelerate their
time-to-answer to reduce overall costs.

Using RF Recording Techniques to Resolve Interference Problems

Radar receiver sensitivity is critical for electronic warfare. The sensitivity of both the radar receiver and the
radar transmitter must be accurate, precise, and repeatable. Sensitivity measurement accuracy directly
relates to power level accuracy — as the radar signal must be strong enough to cover the required
distance. Failure of the radar receiver to decipher signals properly from long distances is not an option for
military applications.

Measurement Challenges
Generally speaking, there are two primary goals of RF interference testing: to ensure interoperability and
compatibility. Interoperability testing focuses on design compliance to a published standard, as well
as margin testing, which helps engineers understand how well a system meets design criteria in the
presence of real-world signal levels and interference. Compatibility testing, on the other hand, focuses on
the “unintended interactions” between a system-under-test and other RF systems.

It’s important for engineers to understand whether radios from different vendors can interoperate with
one another, as well as if all the systems in an RF environment can play together nicely. Ascertaining a
system’s susceptibility to impact from and on other RF assets may also be critical.

To better understand why this approach falls short consider the high-level block diagram of a typical
signal analyzer shown in Figure 22.14. The main limitation to long-duration recording is that test
equipment typically has limited on-board memory.

Display
Preselect Down - Digitize Store Process
Convert
ADC DSP RAM µP
RF Acquire Read &
Input Data Store

Move Data from Digitizer to µP

Figure 22.14. Shown here is a typical signal analyzer block diagram.

Radar and EW Signal Recording and Analysis 261

The signal analyzer does not capture any samples while it post-processes the previously captured data,
effectively creating a gap in its data acquisition. Consequently, if events occur while the previous event is
being processed or if the new event lasts longer than the available memory, it falls into this gap and may
be missed. Moreover, the analyzer’s trigger setup only captures signals for one set of limited conditions.
Once the analyzer fails to capture the event, it is gone forever.

Introducing Gapless Recording

While resolving RF interference problems in complex RF environments can be a tricky task, the gapless
recording offers a viable solution to the measurement challenges presented by the typical signal analyzer.
The technique solves the problem of not knowing when or where an interference event will occur, or how
long it will last, by enabling continuous acquisition of data over long durations. Because there is no gap in
the data recorded, the signal-of-interest, such as an intermittent RF event, is easily captured.

For comparison purposes, consider the data acquisition from a typical signal analyzer with limited on-
board memory, as shown in Figure 22.15. Note the gaps in data that occur once its memory is filled up.

Acquisition Read Acquisition Read Acquisition Read

GAP GAP GAP

FAILURE
MISSED

Figure 22.15. With a typical analyzer, once its memory is filled up, data is “read” from the digitizer to the
microprocessor for processing and display. During this “read,” any new samples available at the digitizer cannot
be processed and are missing, creating a gap in the continuous acquisition of data and resulting in failures being
missed.

Radar and EW Signal Recording and Analysis 262

Now, consider an example of a signal analyzer modified for gapless recording (Figure 22.16). It is the
same signal analyzer shown in Figure 22.14; however, it now includes a high speed data link or bus that
allows the engineer to move data from memory as it is acquired.

Preselect down-convert Digitize Store process

RAM
ADC DSP RAM Storage
buffer
RF Acquire
Input Digitizer PC RAID
data

Figure 22.16. A signal analyzer modified for gapless recording.

Keysight’s gapless recording system is available in predefined packages that have been tested to
guarantee sustained data rates. The configured systems include data interface cards and modules and
can be used with Keysight’s 89600 VSA signal analysis software.

Real-Time Recording and Analysis

Architecting a real-time radar recorder
The aerospace and defense industry investment in threat simulation systems is increasing in line with
the complexity of the electromagnetic battlefield. Most current analysis hardware is limited in capability,
resulting in spot-checking signals using spectrum analyzers or high-end oscilloscopes. In addition, there
is a need to accurately validate the system under test stimulus and output to ensure correct operation
while capturing and recording signals of interest all the time needed to truly validate the proper operation.
Most current analysis hardware is limited in capability, resulting in spot checking signals using spectrum
analyzers or high-end oscilloscopes. Thus, capturing and recording signals of interest all the time is
needed to truly validate proper operation. Keysight’s radar recorder’s capabilities and features are
targeted to solve these problems in the aerospace and defense industry.

Understanding RF system use cases

There are numerous personas with different personalities within the radar industry that exhibit different
areas of focus. RF systems engineers typically know what the outbound pulses look like. Thus, they
are focused on generating accurate pulses and looking for impairments caused by the realities of the
actual hardware. If they find a badly formed pulse, they will need to understand why and how to fix
the component or the subsystem responsible. On the other hand, radar analysts typically collect large
volumes of RF data on a test range and check whether the overall system responded appropriately to an
external threat. If they don’t want their signals to get jammed, they need to know what the signatures are
of different radar equipment by comparing what they are measuring against a catalog of known
pulse trains.

Radar and EW Signal Recording and Analysis 263

In original recorder systems, a workflow associated with the production-oriented testing that the recorder
will witness is needed. Overall lab time was expensive and multiple users competed for time slots, thus
making the most of allocated time with the recorder a priority. In this model, recordings are done in a
batch fashion with recordings piling up as needed for initial analysis and secondary analysis. In a new
recorder system, the system needs the capacity to hold TB’s of data and migrate that data from the
recording portion of the system to the analysis portion as quickly as possible so that some analysis can
be done while the recording is in process as well as in parallel with the next recording.

Figure 22.17. RF recording use cases

There are four main types of RF Recording use Cases: Known/Known = System validation/stress testing,
Known/Unknown = Threat Assessment, Unknown/Unknown = Spectrum monitoring, Unknown/Known =
Difficult test parameters. Today’s focus area is in system validation and stress testing in labs and its main
benefits are that it validates the test source with every pulse and acts as a witness recorder for hardware
in the loop (HITL) testing.

Radar Recorder Use Cases

There are 2 known applications for a radar recorder: emitter validation and jammer validation.
Emitter validation is important because it provides confidence that a platform will perform correctly
in real engagement, outdoor range testing is very expensive, and there is limited time available for
reprogramming. It provides quantitative verification of intentional stimulus. Jammer validation is a
qualitative technique analysis and is also a statistical analysis of input vs output.

Radar and EW Signal Recording and Analysis 264

Emitter validation today is focused on the mission of validating the output of the radar or threat simulation
platform by measuring PRI, pulse widths, amplitude, frequency, and more. Its main purpose is to provide
insight into radar modes of operation and scan patterns.

Jammer validation today is focused on the mission of confirming the expected output of the EW system
under test (SUT) or jammer system. It does this by measuring PRI, pulse widths, amplitude, frequency,
and more and its main purpose is to provide insight into techniques such as pull-off rates, J/S, multiple
false targets, and more.

Validation is important as it provides confidence that a platform will perform correctly during engagement.
Reprogramming is time-limited as test labs have around-the-clock usage and testing is expensive,
especially in outdoor ranges. Thus, the more that can be done in the lab to ensure that the SUT will
perform properly helps ensure proper performance later.

Modern EW threat analysis faces challenges in long simulation time, wideband analysis, and agile
emitters. Thus, Keysight’s radar recorder resolves these challenges by being multichannel and agile.
Threat simulators are inherently multichannel or multiport, measuring AoA is important for scenario
validation, and direction-finding is of interest. These 3 facts support the need for a multichannel radar
recorder. In addition, emitters and threats occur across frequency bands, with communication and radar
signals often occurring at the same time. Keysight’s radar recorder also utilizes absolute timing as there
is a need to analyze time-synchronized events and it is easier to do so when there is an understanding
of when recorded events occurred. In addition, a streaming recording architecture is needed as test
scenarios can range from seconds to hours. Thus, a high rate, gapless data capture system is needed
which is often beyond the capabilities of ethernet.

Capabilities of a Real-Time Radar Recorder

A real-time radar recorder contains the following set of capabilities:

• A multi-channel, integrated RF recorder

• Real-time measurement of pulsed RF signals
• Hours of RF data recording storage per channel
• Real-time scoring of measured pulses to reference definition
• Suite of pulse analysis software

Real-time radar recorders split up these capabilities between various steps that occur while the data
is being acquired and after the data has been acquired. Recording can overlap with data offload
analysis. There is an unspoken time order because data must first be acquired to be analyzed, a series
of recordings can overlap with a series of analyses, thus boosting throughput. In addition, off-system
analysis can be possibly integrated into this process to archive the recorded data. Not every use model
will require off-system analysis, but this is a great feature that real-time radar recorders can provide.

Radar and EW Signal Recording and Analysis 265

Therefore, real-time radar recorders provide multiple levels of analysis ranging from aiding in determining
the current status of a test, visualizing data on the analysis workstation, and offloading data into other
analysis systems.

Real-time radar recorders can be used both as a threat simulation validation tool and as a tool to witness
recording EA system output. These systems allow for input vs output jammer analysis, enable faster time
to confidence in testing, provide the ability to view large amounts of data in a more meaningful way, and
are scalable in size. Thus, these recorders appeal to the two personas in the radar industry mentioned
earlier as the recorder can collect and analyze large amounts of data both during data collection and after.

Keysight’s radar recorder features:

• Selectable bandwidth, and multichannel options

• Real-Time PDW scoring
• PDW domain big data analysis software
• Scalability

These 4 main features together allow for input vs output jammer analysis, enables faster time to answer
and increased confidence in testing, allows the user to view large amounts of data in a more meaningful
way, and has large or mid system applicability.

Exploring Keysight’s range of COTS solutions

Ultra wide analysis bandwidth solutions landscape
Z-Series (63 GHz)

V-Series (33 GHz)

> 4.0
S-Series (8 GHz)

VDI+S VDI+S
Analysis bandwidth (GHz)

(60 to 90 GHz) (75 to 110 GHz)

2.0
VDI+S (50 to 76 GHz)

1.5
Z9071B:
Z9070B: PXA + S-Series (3 to 50 GHz) M1971E + S-Series
1.0 (57 to 90 GHz)
UXA (50 GHz)
0.5

0.16 PXA (50 GHz)

0 10 20 30 40 50 60 70 80 90 100
Carrier frequency (GHz)

Figure 22.18. Determining the best-fit tool depends on the carrier frequency and the required analysis bandwidth

Radar and EW Signal Recording and Analysis 266

Another factor is the number of simultaneous measurement channels needed to accelerate measurement
time. This can be an important consideration when characterizing phased-array radar systems that
contain dozens or hundreds of transmit/receive modules. Figure 22.19 maps analysis bandwidth versus
number of channels for a variety of Keysight oscilloscopes and signal analyzers.

Up 1 1 1
to 4 S-Series V-Series Z-Series
#Channels

PXA
2 2
1 PXA UXA + 3
510 MHz 1 GHz Scope
1.2 GHz

0 MHz 510 MHz 1 GHz 8 GHz 13 GHz 33 GHz ... 63 GHz

BW BW BW BW BW BW BW
RF signal spectral bandwidth
1.  S, V and Z-Series 2 ch full bandwidth, 4 ch half bandwidth.
2.  Up to 50 GHz carrier.
3.  3.6 GHz to 50 GHz carrier.

Figure 22.19. Channel count versus bandwidth is another important consideration when selecting a wideband
measurement tool.

The one-channel PXA signal analyzer can measure signals with carrier frequencies up to 50 GHz
and offers up to 510 MHz of analysis bandwidth. Similarly, the single-channel UXA can handle carrier
frequencies up to 50 GHz and offers a maximum analysis bandwidth of 1 GHz. Note that it is possible to
link multiple PXAs or UXAs to create a multi-channel solution.

An S-Series oscilloscope provides two channels with 8 GHz bandwidth or four channels with 4 GHz
bandwidth. The V-Series oscilloscope offers two channels with 33 GHz bandwidth or four channels with
16.6 GHz bandwidth. The Z-Series oscilloscope offers two channels with 63 GHz bandwidth or four
channels with 32 GHz bandwidth.

An oscilloscope can handle signals with spectral widths nearly up to its bandwidth. One requirement: the
carrier plus modulation must be sampled with enough bandwidth to capture both. For example, a signal
with a 6 GHz carrier and a 2 GHz wide modulation would fit within the 8 GHz bandwidth of an S-Series
oscilloscope and could therefore be evaluated.

Radar and EW Signal Recording and Analysis 267

Utilizing a wideband signal analyzer
The UXA offers a big step forward in usability by providing factory calibration and operational alignments
across the full 1 GHz bandwidth. Integration into the signal analyzer delivers several other benefits: 1 GHz
coverage across the full frequency range (3 Hz to 50 GHz); seamless switching between swept, vector,
and real-time measurements; and easy operation through the streamlined multi-touch user interface (UI).
This combination of capabilities enables informative measurements of pulsed signals in less time
(Figure 22.20).

Figure 22.20. The UXA’s wide integrated bandwidth simplifies the measurement of linear FM-modulated pulses.

A wideband analyzer also offers benefits in measuring narrower pulsed signals. For example, when
measuring pulse rise-times or viewing parameters such as overshoot or droop, the wider analyzer
bandwidth offers a faster sample rate, providing more resolution for these measurements.

It’s also important to see multiple emitters in any radar or EW signal environment. These emitters likely
will also have some intra-pulse modulation and can occur over a wide bandwidth. The UXA can capture
multiple emitters across 1 GHz of bandwidth and calculate statistics over a long period. In addition, the
optional N9067C pulse application automatically detects and analyzes pulses, providing comprehensive
results that enable the full characterization of captured pulses.

Radar and EW Signal Recording and Analysis 268

Using an oscilloscope
For RF designs that need bandwidths greater than 510 MHz or 1 GHz, a digitizing oscilloscope is
an important tool that can assume the role of a “wideband RF receiver.” In these applications, the
Keysight Infiniium S-Series, V-Series, and Z-Series oscilloscopes can be used to make a variety of
FFT and wideband RF measurements, and the input channels are magnitude-flat, phase-linear and
phase-coherent. Analysis capabilities include built-in math programming, and these functions can also
incorporate custom, MATLAB-based programs or algorithms.

The high sample rate and deep memory available in Keysight oscilloscopes make it possible to capture
a radar pulse and extract its envelope information and the embedded coding. Many of the capabilities
built into today’s scopes also support detailed troubleshooting and analysis. For example, the scope
can directly digitize a signal at RF or IF and then process the signal with internal math functions to plot a
histogram or calculate parameters such as absolute value (Figure 22.21).

Figure 22.21. Built-in oscilloscope capabilities make it possible to acquire a chirped radar signal with a bimodal PRI
and then display a histogram of thousands of pulses

Radar and EW Signal Recording and Analysis 269

In radar systems that have more than 300 MHz of instantaneous bandwidth above 63 GHz, the Keysight
N5280A and N5281A downconverters provide four channels of phase-coherent frequency conversion.
With a frequency width of up to 1.5 GHz (N5280A) and frequency range of up to 50 GHz (N5281A), these
devices can down convert ultrawideband radar signals into the direct digitizing range of Infiniium series
oscilloscopes. When system characterization requires time-correlated measurements between the analog
and digital domains, a configuration that includes a Keysight mixed-signal oscilloscope (MSO series)
provides an easy, efficient solution.

In the design of radar and EW systems, increasingly complex pulse compression techniques are being
deployed to maximize resolution and range or to reduce the likelihood of detection. To help you identify
and measure performance, test solutions must deliver high resolution, excellent dynamic range, and wide
analysis bandwidth. Keysight’s range of COTS solutions include vector signal analysis software, wideband
signal analyzers, and multi-channel wideband oscilloscopes. Compared to traditional approaches,
these tools provide enhanced ease of use that simplifies the process of producing accurate, repeatable
measurement results.

Simplifying the Characterization of Wideband Pulsed Signals

Gaining simplicity through integration
The UXA offers a big step forward in usability by providing factory calibration and operational alignments
across the full 1 GHz bandwidth. Integration into the signal analyzer delivers several other benefits: 1 GHz
coverage across the full frequency range (3 Hz to 50 GHz); seamless switching between swept, vector,
and real-time measurements; and easy operation through the streamlined multi-touch user interface (UI).
This combination of capabilities enables informative measurements of pulsed signals in less time
(Figure 22.22).

Figure 22.22. The wide integrated bandwidth simplifies measurements of linear FM-modulated pulses.

Radar and EW Signal Recording and Analysis 270

In contrast, the two-instrument approach (i.e., downconverter and digitizer) requires the user to run a
separate calibration with an external source or reconfigure connections to get a flat, usable band or
uses a generic correction that provides less accuracy. Additionally, the wideband capability cannot be
accessed in low band with the two-instrument approach, limiting the minimum frequency to at least
several gigahertz.

Using the broader applicability of wide bandwidth

While another measurement receiver is needed when a signal occupies more bandwidth than the
analyzer, it may be less obvious that a wideband analyzer offers benefits in measuring narrower pulsed
signals. For example, when measuring pulse rise-times or viewing parameters such as overshoot or
droop, the wider analyzer bandwidth offers a faster sample rate, providing more resolution for the
measurement. As shown in Figure 22.23, this can be an important aid to understanding the signature of
the waveforms.

Figure 22.23. The image on the left uses half the sample rate of that on the right. The larger sample rate (or wider
bandwidth) shows a clearer envelope of the signal and hence a better representation of the signature of the pulse.

Radar and EW Signal Recording and Analysis 271

Marker resolution is also improved with a wider bandwidth. Other techniques such as zero-padding can
also be used: this will help you see the signal; however, it will still mask the true signature of any fast-
moving transient activity such as overshoot.

Finally, it’s important to see multiple emitters in any radar or EW signal environment. These emitters likely
will also have some intra-pulse modulation and can occur over a wide bandwidth. The UXA can capture
multiple emitters across 1 GHz of bandwidth and calculate statistics over a long time. The N9067C
embedded pulse application automatically detects and analyzes pulses, providing comprehensive results
that enable the full characterization of captured pulses (Figures 22.24).

Figure 22.24. The N9067C embedded pulse application can display numerous pulse statistics, simplifying the
characterization of dense pulse environments.

Radar and EW Signal Recording and Analysis 272

Applying the flexibility of a complete solution
Combining these functions into a single instrument enables engineers to scan gigahertz of the spectrum
with swept analysis, see dynamic multi-emitter signal activity in real-time, and make detailed wideband
pulse measurements—all in less than two touches, saving time whether measuring or troubleshooting
(Figure 22.25).

Figure 22.25. The large multi-touch display provides side-by-side views of real-time measurements and tabular
pulse metrics, which can be easily exported for post-analysis.

In the UXA, an enhanced 255 MHz bandwidth path co-exists with the 1 GHz wideband path, offering
approximately 80 dBc of dynamic range and gap-free RSTA capability. In addition to ensuring that no
signal is missed, real-time also provides frequency-mask and time-qualified triggers (Figure 22.26). Users
can prune and select specific signals in a dense environment using time, frequency, and amplitude
triggering, or any combination of the three.

Figure 22.26. Frequency-mask trigger enables frequency-selective analysis with RTSA up to 255 MHz of bandwidth.

Radar and EW Signal Recording and Analysis 273

The X-Series signal analyzers are the benchmark for accessible performance that puts you closer to the
answer by easily linking cause and effect. Across the full spectrum — from CXA to UXA — you’ll find the
tools you need to design, test, and deliver your next breakthrough.

With wide analysis bandwidth, the flagship UXA delivers wide-open performance and deeper views of
elusive and wideband signals. In-depth analysis is made easy with the pulse application software and the
14.1-inch screen with multi-touch UI. With its familiar spectrum-analyzer user experience and capabilities
never offered in an integrated instrument, the UXA enables you to see more and take your designs farther.

Radar and EW Signal Recording and Analysis 274

Chapter 23
Electronic Intelligence
Electronic Intelligence (ELINT) is a subdivision of Electronic Support (ES) within Electronic Warfare
(EW). Elint is concerned with the non-cooperative interception of electromagnetic signals that do
not carry communications. Such signals cover all categories of radar, beacons and transponders,
jammers, missile guidance, radio altimeters, navigation emissions, and identification systems
(IFF). This book is primarily focused on testing the serviceability and the efficacy of friendly radar
systems and the ability of friendly Electromagnetic Attack (EA) to be effective against radars of
other countries.

Efficacy is concerned with effectiveness of a radar’s ability to withstand an electromagnetic attack

(EA) or to be able to mitigate such an attack. While strictly not a radar signal, identifying and
characterizing EA signals is also an essential task. EA encompasses jamming, spoofing (false target
generation), and intelligent attack (range gate stealing, eg).

It is important to note that the scope of EA is wider than jammers alone as many new systems are
able to use machine intelligence to generate a sophisticated multi-faceted attack with the objective
of fooling the radar and not merely jamming it. This new approach to EA is in response to new low-
probability of intercept radars, radars that utilise spread spectrum techniques, smart radars using
beam management to mitigate EA and multi-static radars.
Spectrum scanning and analysis for both radar signals and EA signals requires similar technical collection
techniques and will include:

• Frequency or frequencies of radar or EA signal

• Pulse width and pulse shape
• Pulse repetition frequency
• Power radiated calculated from the radar equation
• Pulse compression codes (barker, costas, etc)
• Frequency spectrum from frequency shift keying
• FMCW characteristics
• Radar or EA beam characteristics
• Frequency agility profile

Technical parameters collected are aligned with signatures and are stored in a Parametric Database.
Radar and EA parameters are collected under a Pulse Descriptor Word (PDW) and are assigned to a
signature. An integrated collection of Signatures is the basis for an Electronic Order of Battle (EoB).
Electronic Support (ES) military units (aircraft, UAVs, satellites, maritime units, and terrestrial army units)
revisit signatures periodically to confirm or update. Occasionally new objects are found and after analysis
new signatures is added. Some parameters are gathered through other intelligence channels not involving
the electromagnetic spectrum.

Electronic Intelligence 276

Important intelligence assets may be subject to continuous monitoring (new or strategic systems, eg).
In times of conflict the EOB is central to warfighting in the electromagnetic spectrum, and together with
cyber intelligence forms the core element of Cyber & Electromagnetic Activities (CEMA). All friendly
electromagnetic assets must comply with the CEMA strategic plan (see Figure 23.1) and hence testing of
these assets employing PDWs (signatures) is mandated.

Commanders, supported by their staffs, must integrate and synchronize cyberscape

operations, electronic warfare, spectrum management operations, and related
capabilities to achieve the desired effects in support of unified land operations.

Cyberspace Operations
Employ cyberspace capabilities
to achieve objectives.

• Offensive cyberspace operations

• Defensive cyberspace operations
• DOD Information network operations

Cyber
Electromagnetic
Electronic Warfare Activities Spectrum
Use electromagnetic and
Management
directed energy to control Operations
the electromagnetic spectrum
or to attack the enemy. Plan, coordinate, and manage the
use of the electromagnetic spectrum
• Electronic attack through operational, engineering,
• Electronic protection and administrative procedures
• Electronic warfare support to deconflict all systems.

Figure 23.1. CEMA Integration (US Army 2014)

Electronic Intelligence 277

PDWs are used to set up testing environments for both radar and EA. For radar, a model test EA
system is formulated and used to test the efficacy and effectiveness of a radar to operate under attack
conditions. Figure 23.2 represents a radar test environment for EA scenarios.

Radar under Test Electronic Attack Emulator

RF Rx Spectrum Analyser
Transmission Electronic Attack Pulse Descriptor Words
Emulator Scenario Generation (PDW) from ELINT
Radar Signal Generator

Noise Generator

Range deception
Radar Mode
Operation
Velocity deception
Radar
Operations
Analysis

Figure 23.2. Radar test environment under jamming attack

For the effectiveness of a friendly EA asset, a radar test environment is set up using PDWs representing
an enemy radar as shown in Figure 23.3.

Radar Emulation
Scenario Controller
Channel Emulation - PDW Domain Radar Echo Generator

Tx Emulation

Rx Emulation
Channel Emulation - PDW Domain Signal Generator

Radar Rx Front End

Signal Accumulator (AGC) Emulation
Wideband
Digitizer
Sysytem Under Test
Channel Emulation - Time Domain Signal Conditioning

Figure 23.3. EA test environment against enemy radar

Electronic Intelligence 278

Chapter 24
Test System Calibration and Alignment
When generating a realistic scenario capable of testing the performance of a modern SUT, some
key concepts must be utilized.

Initially, we need to look at the system calibration process. First, there is the need for system
coherence. Next, we need amplitude alignment, phase alignment, and time alignment to emulate
angle of attack, also called angle of arrival (AoA). Once we have a coherent system that is
amplitude, phase, and time-aligned, we can use the calibrated system to create a realistic
emulation of an electronic warfare electromagnetic environment.

Accurately modeling the environment of an electronic warfare scenario requires generating the
signals that represent the angle of arrival characteristics of all platforms of interest. To make the
emulation of such signals meaningful there must be a very careful creation of amplitude, phase, and
timing differences between signals as would be received on multiple antennas of the SUT. The first
step in this process is to remove as much error in the signal generation system as possible. This is
the primary purpose of calibrating an EW emulator.

Realistic Model
Amplitude of Electronic
Time Alignment
Alignment Warfare
Environment

System Phase Calibrated

Coherence Alignment System

Figure 24.1. System calibration process

Test System Calibration and Alignment 279

Create a Successful Model by Analyzing Considerations
As the threat emulation environment is built up, one of the first considerations needs to be the time-base
of the system; specifically, all emitters need to be phase-locked to the same time base. A fully coherent
system is critical for a successful model of threat emulation.

Phase coherence is a critical consideration to ensuring success. Two systems are coherent if they have
a constant, relative phase at all instances in time. Signals can have phase noise and phase drift and still
be coherent. When the phase impediments are common to both signals, they will cancel out and not be
present in the relative phase. It can be seen by how coherence is described in the given equation.

coherence
expected value operator
average value
std deviation
covariance of signal X and signal Y
signal X
signal Y
Equation 24.1. Coherence equation

Signals with the coherence of value 1 are fully coherent and those with value 0 are completely non-
coherent. When the coherence is somewhere in-between with partial coherence, they are called phase-
stable signals.

A second factor, besides needing phase coherence, is having signal alignment. This is when the
relative amplitude, phase, and time can be set to near 0, and then adjusted to desired values. Through
calibrations, one can know what the relative amplitude, phase, and time are as a starting point. Then,
differences in amplitude, phase, and time between the signals can be introduced to emulate accurate
AOA characteristics.

Three Dimensions to Calibrate On

1. Amplitude

2. Phase

3. Time

The main dimensions to calibrate on are amplitude, phase and time. This can be done at a single
frequency point or across multiple frequencies depending on the SUT design requirements. This removes
as much uncertainty from the system as possible.

Test System Calibration and Alignment 280

Amplitude
The difference in the amplitude of two signals in a system defines the amplitude alignment. Some SUTs
determine AoA through the amplitude difference in signals received on multiple antennas, so calibrating a
threat emulation signal generator for accurate signal amplitudes is important.

Az

+ Az

Figure 24.2. Amplitude alignment diagram

Equation 24.2. Amplitude alignment equation

Phase
The difference in the phase of two signals in a system defines the phase alignment. Once signals are
phase aligned, they can be programmed for desired phase shift.

Equation 24.3. Phase alignment equation

Incident
wavefront

Phase
difference

d
Antenna Antenna
1 2

Figure 24.3. Phase alignment diagram

Test System Calibration and Alignment 281

Time
An aligned system is defined by the difference in time between the two pulse envelopes being nearly zero.
Desired envelope time shifts to emulate AoA characteristics can be programmed.

Wave speed & direction Wave front that has

passed the receivers

Origin at Receiver P0

z
V
P0 Y
Pm
X
R0
PN Rm
RN

R1 E
P1
Emitter

Receivers
Wave front
Approaching the receivers

Figure 24.4. Time alignment diagram

Equation 24.4. Time alignment equation

Once the threat emulator is coherent and the output signals are aligned, RF threat emitters can be
accurately emulated. Amplitude, phase, and time alignment in a system allow for accurate calibration
between the emulator and the threat, thus ensuring that your threat emulator is accurate. Keysight’s
89600 VSA software can be used with Analog and Vector UXGs to create an accurate threat emulator.

Test System Calibration and Alignment 282

Chapter 25
Field Testing, Sustainment, Uptime-Service, and Support
The typical aerospace/defense (A/D) weapons programs can last decades, hence a consistent and
reliable performance must be ensured. The sustainment phase of an A/D program makes up most
of the program’s lifetime cost, up 70% of the total program cost1. However, it is underappreciated
and undervalued when key decisions are made during the early stages of the program2.

These milestone decisions made during the early stages incur upfront costs that are relatively
high compared to the rest of the program’s per-incident costs. However, these key decisions
also define the rest of the program, including the long-term costs during the sustainment phase.
For example, downtime due to improper planning of test equipment maintenance, especially
during production runs, can put a program budget and schedule at high risk. It is therefore vital to
consider sustainment challenges and costs in the early stages of the program, and the associated
requirements and consequences for program suppliers.

Field Testing, Sustainment, Uptime-Service, and Support 283

Sustainment Definition and Requirements
How the aero/defense community defines sustainment
The term “sustainment” is used to refer to many different applications, based upon the varying
perspectives of roles within the aerospace/defense ecosystem. For example, a government program lead
has a different goal for sustaining the program compared to the government contractor’s perspective and
goals. Similarly, a test equipment supplier often talks about sustainment concerning long-term support of
test equipment hardware. Though supporting test equipment long-term does influence overall program
sustainment costs, it is normally not the end user’s main concern. Therefore, it is necessary to consider
the end user’s perspective of sustainment in relation to overall program cost and success.

Let’s consider the concept of sustainment according to the government. Reference 3, for example,
lists several key goals for sustainment activities. The first is to maximize the chance of mission success
through guaranteed availability to meet mission requirements. This is measured through parameters
that quantify operational availability and the probability of success of the given mission. Second, unit
self-sufficiency during the mission is a goal that is measured by the length of time a unit can maintain
self-sufficiency. In this context, self-sufficiency means that a combat unit can operate without external
resupply or support for a given period or combat pulse. Self-sufficiency must be balanced by the total
number of personnel required for a given combat unit to be successful. The goal is to minimize the
number of additional forces dedicated to sustainment while maximizing the length of time a unit can be
self-sufficient.
Thus, you can see how sustainment is an essential component of mission success. In addition, expenses
due to ongoing operational and maintenance activities are becoming an increasingly larger part of
total expenditures. This is shown in Figure 25.1, which depicts total expenditures by category for A/D
programs2. In the chart, the operations and maintenance category has increased by 10% from the
1980s to the 2000s. It is projected to increase further to 44% of total expenditures by 2029. Personnel
expenditures have also increased, likely due in part to increased demand to meet sustainment goals.
Unfortunately, although sustainment costs are becoming more significant, it continues to be undervalued
and underappreciated during the key decision phases.

Expenditures by category
2%
7% 4% Other
9%
Research, development, testing, and evaluation
11% 13%
Procurement
18%
Personnel
20%
Operations and maintenance
30%

27%
24%

23%

39% 44%
29%

1980-1989 2000-2009 2029e

Figure 25.1. Total expenditures by category for A/D programs

Program sustainment and consequences for suppliers

Development Initial Full
RFP Operational Operational
Full Rate Capability
CDD Release Production Capability
Decision (IOC) (FOC)
Validation (FRP) Decision
Material
Development
Decision

A B C
LRIP
Operational Test and
Evaluation (OT&E) Sustainment Disposal
Material Technology Engineering & Production & Operations & Support
Solution Maturation & Manufacturing Deployment
Analysis Risk Development
Reduction

Legend: = MIlestone Decision = Decision Point

Figure 25.2. Classic program model for a major weapons platform

Now that we have established what the goals are for sustainment and that it is becoming increasingly
important to manage sustainment-related expenses, let’s consider the program support strategy and
the associated consequences for the suppliers of those programs. Figure 25.2 depicts the program

Field Testing, Sustainment, Uptime-Service, and Support 285

model of a hardware-intensive A/D program such as a major weapons platform4. This is taken from the
DoDI 5000.02 procedures. Enclosure 6 from the DoDI 5000.02 document highlights the requirements
for life-cycle sustainment for such a program. From the first key decision stages, the product support
strategy must be developed. This support strategy is the basis for all sustainment efforts to maintain the
requirements of the program.

Included in the product support strategy are:

• Sustainment metrics that are continually monitored and managed to minimize program costs and
risks to schedule and quality
• An initiative to improve reliability based upon failure modes and critical analysis. Data is captured
during the engineering development phase and through systems health info that is available through
on-board and off-board technologies
• The ability for competition at the prime and subcontract levels for both system and subsystem levels

The sustainment plan is not fixed. Throughout the lifecycle, the plan performance will be assessed and
modified as necessary to reduce cost and risk.

What, then, is the impact on suppliers? Ongoing assessment of a supplier’s ability to provide support is
part of the overall management of sustainment for the program. It is therefore imperative for the supplier
to provide products and product support for the length of the program. It also means that consistent
yields are required. In addition, for the supplier to continue to provide the lowest possible cost and remain
competitive, predictable costs are required. For example, if the supplier’s costs are at risk of increasing
over time, this increases prices for long-term programs and reduces the supplier’s competitiveness. And
finally, to improve sustainment metrics and manage long-term costs, contracts are often re-bid at later
program stages. When a supplier can maintain consistent costs and continue to improve efficiency, the
chances of winning the re-bid are significantly higher.

Current efforts to support program sustainment

Consider which trends are driving the challenges of sustaining long-term programs and the actions
that program contractors are taking to mitigate these challenges. Increasingly, program contractors are
focused on outsourcing unnecessary tasks and using more commercial, off-the-shelf (COTS) equipment.
This eliminates the need for the contractor to spend time and resources on non-core activities such
as tests, which, in turn, enables the contractor to focus on core expertise, driving both innovation and
competitiveness. In addition to outsourcing, programs are looking at strategies for reducing total lifecycle
costs rather than immediate expenditures. Also, the business model is moving from cost-plus to fixed-
price, meaning budget controls are becoming more essential for ensuring program profitability.

Sustainment requirements are addressed by program contractors through:

• Implementing a zero-defect policy for suppliers to ensure uncompromising quality; this means that
suppliers’ parts must always meet their specifications
• Minimizing downtime to mitigate schedule delays

Field Testing, Sustainment, Uptime-Service, and Support 286

 • Demanding consistent performance and yields while the margins between instrument performance
and product specifications continue to decrease

As a result, understanding the limits of test system performance and maintenance is essential for
sustainment purposes.

Taking a Total Cost of Sustainment Approach

Key challenges and risks for long-term programs
Based upon the definition of sustainment efforts and requirements, it is possible to identify some of the
key risks associated with maintaining long-term programs. First, programs are not always fixed for life. At
multiple stages, re-bids put profits and all previous work on the program at risk. Therefore, maximizing
the probability of a successful re-bid is essential. Second, maintaining the lowest possible costs ensures
profitability. However, it is important not to focus solely on upfront costs. When accounting for long-term
program costs, it is important to account for the total cost of sustainment for the life of the program.
Third, downtime, planned and unplanned, puts a program budget and schedule at high risk. Having a
flexible and dynamic solution for uptime support is therefore essential. On the other hand, unplanned
downtime is often a significant issue. Not putting a plan in place, such as acquiring spares, can put the
schedule at serious risk. Still, having spares available may not be enough. If spares are not properly
maintained and tracked, they may not be usable when needed. Or, even worse, if that spare is not being
calibrated at regular intervals, it may appear to solve the immediate challenge of a broken instrument but
in fact, be contributing to yield problems.

Finally, support costs become significant when a commercial product is reaching its end of life.
Unfortunately, not only are costs higher at the product’s end of life due to shortages in parts supplies, but
failures tend to occur more often. These risks are more apparent for long-term programs because the
lifecycle of most commercial products is shorter than that of most A/D programs.

An underlying issue becomes clear when analyzing the various challenges of sustaining long-term
programs: all these risks require time and effort to manage and mitigate. The real goal, then, is not just
to mitigate these risks. Instead, it is to eliminate the need to think about them at all. If all these risks are
addressed by a third party with the appropriate expertise, then a defense contractor can truly focus on its
core proficiencies, and this in turn increases its competitiveness, profitability, and longevity in the market.

Field Testing, Sustainment, Uptime-Service, and Support 287

Total cost of sustainment example
Consider Figure 25.3, which depicts two different methods for addressing program sustainment and
costs. The first is the traditional approach, which will be called the per-incident method. In general, this
will save the most money at the beginning of the program because all support costs are paid per incident,
and support is typically required less often. In this case, support costs include everything required
to maintain the system: repair, calibration, training, etc. This can also include the cost of buying and
maintaining spares or periodically paying for onsite services if needed. The other method is equivalent to
paying for an annual agreement that covers all required support. A single annual price is charged by the
provider of support services, and this price increases each year by only the inflation rate. With a long-term
agreement, the service provider can plan, control costs, and ensure a consistent price even when repairs
and other services are required more often at the end of the program.

$500,000 Total Cost of Sustainment Over Program Life: PNA-X Example

Total Cost of Sustainment

$450,000
$400,000
$350,000
$300,000
$250,000
$200,000 Per Incident Method Agreement Method
$150,000
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Major Major Time between major repairs

Annual Expenses

Recurring Hypothetical repair repair decreases toward end of life

expenses Major end of support
repair
Opeartor
turnover

Yr1 Yr2 Yr3 Yr4 Yr5 Yr6 Yr7 Yr8 Yr9 Yr10 Yr11 Yr12 Yr13 Yr14 Yr15 Yr16 Yr17 Yr18 Yr19 Yr20 Yr21 Yr22 Yr23 Yr24 Yr25

Program Life (Years) ANNUAL EXPENSES (PER INCIDENT)

Figure 25.3. Total cost of sustainment example for a network analyzer

The upper plot in Figure 25.3 shows the total cost of sustainment as a cumulative total from years 1
to 25. For simplicity, the figure shows the repair and calibration costs for a Keysight vector network
analyzer. The lower plot in Figure 25.3 shows the annual support costs for the per-incident method. The
agreement price grows at a 3% inflation rate, so it was not included in the lower plot. Until about year 5,
the two methods have similar prices (red and gray lines). But, the cost of each major repair is significantly
higher than the agreed price. In general, the agreed price for the entire year can be as much as 70%
lower than the per-incident price of one major repair. Therefore, by year 10, the agreement method
becomes the lower-cost option. As the program reaches its long-term sustainment phase, the associated
equipment has aged and tends to fail more often, decreasing the meantime between major repairs. This
creates a major difference in total cost by year 25, the end of the program. In this case, the total cost of
sustainment of the agreement method is only 66% of the total cost of paying per incident.

Furthermore, it is important to note that it does not take many failures to justify using the agreement
method. To maintain an approximately equal total cost between the methods, no more than two major
repairs can occur within 25 years.

Field Testing, Sustainment, Uptime-Service, and Support 288

A consistent cost is a winner when it comes to achieving the lowest total overall cost. Predictable costs
help in other ways, too. For one, knowing the cost to maintain a test system for the entire program life
means the true costs are already known if a program must be re-bid. This makes a significant difference
compared to the risks of the per-incident method. As insurance, the re-bid is often padded to account for
the per-incident costs that may arise. This can significantly reduce the chances of winning the re-bid. This
example shows that extending the life of the instrument using the agreement method provides the lowest
total cost.

An Innovative Approach to Sustainment

Ensuring predictable costs for long-term support
Consider a test system made up of many elements, including instruments, components, signal routing,
and switching elements, and control and computation elements such as CPUs, etc. This normally
includes elements from many manufacturers, as well as custom-designed components from third-
parties or by the program contractor itself. Every element of the system will be at a different point in its
lifecycle. In addition, each element’s lifecycle may be shorter or longer based upon the type of element
and its associated demand cycle, parts availability, industry’s pace of evolution, etc. One approach used
by OEMs that is normally expensive and usually cost prohibitive is to look at expected failure rates for
all components in the system and stockpile enough spares to cover the entire life of the program. This
creates an enormous cost upfront and does not consider the evolving failure rates and requirements of
the program itself.

An alternative approach is possible and would dramatically reduce upfront costs while managing
ongoing expenses appropriately. Using this approach, the program contractor partners with the OEM to
plan for long-term support. An end-user who chooses to adopt this approach can achieve predictable
sustainment costs because the OEM using this planning approach can ensure consistent prices. This
partnering approach utilizes a key role that must be created by the OEM called the “project success
manager.” The project success manager is responsible for ensuring that support is available, provided
only as needed, and delivered at the lowest possible cost, which translates to a low, predictable cost for
the program contractor.

There are four key aspects of this approach that are implemented by the project success manager.
First, for system elements that are manufactured by the OEM service provider and have not been
discontinued yet, the project success manager works with product planners to ensure that parts are
available throughout the life of the program. Product planners then make lifetime buys when necessary
to ensure that parts are available for standard support purposes and any existing sustainment contracts.
Second, the project success manager continually audits the system bill of materials to determine when
obsolescence is imminent and secures the appropriate parts or spares. For example, a CPU often has
a much shorter lifecycle than a typical A/D program. The project success manager must ensure that
enough spare CPUs are available, which are compatible with the system and meet program requirements.

Field Testing, Sustainment, Uptime-Service, and Support 289

 As a system element approaches obsolescence, the correct approach for managing parts varies based
upon the type of component. In many cases, it is necessary to stockpile parts accordingly. The availability
of parts depends heavily upon where the system element is in its lifecycle, the manufacturer’s strategy
for supporting elements, and the overall demand for that element’s parts and components. The third task
implemented by the project success manager is to ensure that each element has enough supply of parts
using any means available. A system element that is past its obsolescence period, for example, may be
only available at a marketplace for used equipment. Or, due to the risk of counterfeit parts, there may not
be any parts available from trustworthy sources. In this case, utilizing in-house expertise in component-
level repairs can ensure parts are available for decades beyond obsolescence.

The final aspect of this novel approach for sustaining test systems is to continually plan, especially
earlier in the program’s lifecycle. The project success manager, therefore, ensures that predictable,
consistent costs for support are achieved. This translates to an overall predictable cost of sustainment
for program contractors.

When to plan for sustainment?

When is the best time to start planning for sustainment? As discussed, costs can be reduced when
appropriate planning occurs in advance of the product obsolescence period. However, in most cases, it is
possible to still achieve significant value and savings no matter where the program is in the cycle. At least
part of the issue is mitigated by the fact that few if any of the products are on the same cycle. Many of
the newest products will likely be free of parts shortages for the majority of the program. For any products
built with parts that may be in short supply in the near term, possible solutions include a decision to use a
combination of used equipment, parts stockpiling, and component-level repair capabilities.

Consider Figure 25.4, which depicts the tradeoff between the value and cost of a sustainment solution
based upon which program phase is currently employed. At a high level, this graphic depicts the inverse
relationship. The earlier that you plan for sustainment, the more options there are available to achieve the
lowest cost. One reason for this, as discussed, is the number of options available to acquire or supply
stockpiles of parts. In addition, besides parts supply, sustainment and test costs can be significantly
reduced through a combination of services that are tailored according to the program phase and goals.

Greates ly
t flexibil est supp
ity, lowe cost, low
Cost provide

highest
Flexibility, value

sustainment

st cost,
most va
lue Least flexibility,

RFI/RFP/RFQ Award/Protype Production Deployment / Modernization / Decommission

Figure 25.4. Tradeoff between value and cost based upon program phase

Field Testing, Sustainment, Uptime-Service, and Support 290

For example, in the definition phase, consulting to help design the test system and improve test
processes reduces long-term costs. Later in the lifecycle, there is still time to do significant planning.
Onsite services can be utilized, such as providing a dedicated expert to ensure maximum uptime when
it is needed and for critical systems that need it most. Better management of assets through available
programs and tools can help identify utilization and health of instruments, producing insights into reducing
capital expenditures, operational expenditures, or both. By performing analysis regarding the timing of
obsolescence, actionable insights can be used to determine which instruments are candidates for a
technology refresh and the best timing for that process.

Comprehensive Uptime Program

Planned and unplanned downtime must be addressed in addition to long-term support to minimize
ongoing costs. It is important to understand the cost and benefit tradeoffs of any downtime mitigation or
uptime program, from a simple spare strategy to a comprehensive partnership with an OEM. Also, note
that the optimal uptime strategy may evolve as the program moves through its lifecycle stages.

To better understand the cost versus benefit tradeoff, consider the cost of downtime according to the
current program stage. For example, in the earlier development and prototyping stage, downtime is
more manageable and may be mitigated through a spare program or even a return-to-OEM strategy with
dedicated turn-around-times. On the other hand, when a system is getting used regularly to identify and
verify new threats, the cost of downtime is likely much higher. This is especially the case when lives are
at stake. The difference between sending individual instruments away to be calibrated and an on-site
approach can reduce downtime from months to days, or even hours. Check with the original equipment
manufacturer (OEM) on available solution-level services that reduce downtime. Examples could include a
customized solution of on-site repair and calibration services, dedicated loaners for any instruments that
cannot be serviced on-site, and guaranteed response times for on-site service calls.

Consider the following case study that highlights how partnering with the hardware solution provider can
mitigate program and schedule risks while providing consistent access to the system. In this case study,
the defense contractor needed to increase its system output but did not have the people, equipment, or
floor space to meet the requirements. Current system uptime was less than 70% and was considered
acceptable based upon uptime measured on other systems and programs. However, by partnering with
the OEM system provider on a comprehensive uptime program, the contractor achieved consistently
greater than 90% uptime. This additional uptime translated to greater capacity, which allowed the
contractor to achieve the increased production targets without acquiring more facilities or floor space. The
comprehensive program consisted of OEM expert personnel embedded in the contractor’s environment
with a working set of spares and parts to provide all required services inside the secure environment. All
moving of parts and equipment in and out of the secure environment was performed independently, and
therefore this did not impact the contractor’s system uptime.

Field Testing, Sustainment, Uptime-Service, and Support 291

Of course, it is important to understand, based upon program requirements and system usage, the
impact to cost and schedule that any of these solutions will have relative to the cost of each potential
solution. In the case study, the defense contractor achieved such a large benefit with increased
capacity that it far outweighed the cost of embedded OEM personnel dedicated to servicing the
contractor’s systems.

Technical Support
Technical support is a key component of the support plan for any test system. Understanding the use of the
system, or perhaps even more importantly, identifying whatever is causing unexpected issues with the test
system is an absolute necessity to most efficiently utilize the time spent on the system and increase overall
capacity. This ensures schedules are met and system use is maximized. Technical support at the system
level can take on many forms, whether remote or on-site. Based upon the program stage and requirements,
it is important to consider the best method for receiving regular technical support, as well as the process
that is used in the event of a support issue. For example, while on-site technical support may be especially
useful in the beginning after the system has first been delivered, remote support may be sufficient once
the initial learning phase is completed. While a remote expert would not be able to see the system, he or
she should know about the system architecture and would still be able to troubleshoot issues remotely. In
addition, any required service such as a repair could be quickly identified and ordered remotely.

Sustainment is a major contributor to the overall program cost, and in many cases is the leading
contributor. This has consequences. For example, long-term supportability is required. Also, it is
important to maintain predictable costs and utilize cost-management techniques (e.g., improve efficiency).

Not only is sustainment a leading contributor to the program cost, but most of these costs are also
locked in by decisions made during the development phase. As shown, managing risk through a planned
approach to sustainment can ensure predictability and lower overall costs versus addressing and
managing the issues when they arise. Predictable costs ensure maximum probability of success when
re-bidding contracts and maximum probability of total program profitability. They also remove the risk of
high-cost per-incident support events.

Unfortunately, it has been common to accept the risk of managing issues when they arise. This may seem
to be the lowest-cost option, especially at the beginning of the program when support costs are usually
low. However, this is not the best long-term option. The costs can be very high, and there is no way to
predict the magnitude or timing of those costs as the program ages if no planning has been done earlier.

Predictable and consistent costs are achieved by partnering with an OEM that is committed to delivering
the right level of support throughout the lifecycle. The innovative approach to sustainment solutions
discussed ensures test systems are operational and maintained long-term, provides the right level of
uptime support, and minimizes total sustainment costs. Additionally, while incorporating sustainment
solutions earlier in the program enables greater value and cost savings long-term, a customized
sustainment solution can be created for any phase of the program and set of requirements. One example
of this is an evolving uptime solution tailored to the given program stage and requirements.

Field Testing, Sustainment, Uptime-Service, and Support 292

Ultimately, by leveraging a customized solution for sustainment, the program contractor eliminates
the excess time and resources previously required and can focus on true core competencies and
differentiators. Thus, profits are maximized, a schedule is maintained, and long-term success is achieved.

References
1. D. Gouré, “Pentagon Plan To Manage F-35 Sustainment Is A Good Move,” lexingtoninstitute.org,
Jun. 1, 2015. [Online]. Available: https://www.lexingtoninstitute.org/pentagon-plan-to-manage-f-35-
sustainment-is-a-good-move/. [Accessed: Feb. 12, 2019].
2. S. Hurt, “Three Ways the U.S. Department of Defense Can Achieve Its Sustainment Objectives in
Challenging Times,” atkearney.com, Feb., 2013. [Online]. Available: https://www.atkearney.com/
aerospacedefense/article?/a/three-ways-the-u-s-department-of-defense-canachieve-its-sustainment-
objectives-in-challenging-times. [Accessed: Feb.15, 2019].
3. E. Peltz, “Equipment Sustainment Requirements for the Transforming Army,” rand.org, 2003. [Online].
Available: https://www.rand.org/pubs/monograph_reports/MR1577.html.[Accessed: Jan. 18, 2019].
4. Executive Services Directorate, “Department of Defense Instruction 5000.02,” Executive Services
Directorate, Aug. 10, 2017. [Online]. Available: https://www.esd.whs.mil/Portals/54/Documents/DD/
issuances/dodi/500002_dodi_2015.pdf. [Accessed: Feb. 2, 2019].

SUSTAINMENT AND UPTIME SOLUTIONS FOR LONG TERM AEROSPACE

DEFENSE PROGRAMS
https://www.keysight.com/see/radar-ew-system-test

Field Testing, Sustainment, Uptime-Service, and Support 293

Chapter 26
Software Testing for Radar & Electronic Warfare Systems
In addition to testing the hardware components of radar and electronic warfare systems, growing
importance is attached to testing the software of these systems. This chapter will highlight some of
the developments and emergent best practices of software testing.

Software Testing for Radar & Electronic Warfare Systems 294

Software Test Automation and the Development Lifecycle
Software test automation is a key component of DevOps. Recent years have seen a push towards
DevOps across IT systems in both private industry and national defense. DevOps refers to the set of
practices that automate processes between software development and operations. The goal of DevOps
is to build, test, and iteratively release more reliable software in a faster manner, often involving and
complementing Agile methods.

Emphasizing the importance of security, the concept of DevOps is extended to DevSecOps. The US Air
Force’s Office of the Chief Software Officer defines this as “the software automated tools, services, and
standards that enable programs to develop, secure, deploy, and operate applications in a secure, flexible,
and interoperable fashion.” Among the benefits of the adoption of DevSecOps are shortened release
cycles, higher-quality releases that can incorporate continuous feedback from real end-users, and the
automation of testing and security.

Software Test Automation is a Requirement of DoD Projects

Traditionally, software testing has been conducted through the writing and execution of test scripts.
Automated testing allows teams to automate repetitive, rule-based tasks to accelerate testing and
widen test coverage. In October 2020, the US Department of Defense published its latest guidelines on
software test automation in DoD Instruction 5000.87, Operation of the Software Acquisition Pathway.
The Instruction sets out the policy and procedures for “efficient and effective acquisition, development,
integration, and timely delivery of secure software.”

The iterative development approach encapsulated by the phrase “Software is never done” suggests
a constant cycle of planning, coding, building, and testing. More specifically, continuous testing is
a requirement of the execution phase of software development, with DoDI 5000.87 indicating that
whichever software development methodology is being used, it must “incorporate continuous testing and
evaluation, resiliency, and cybersecurity, with maximum possible automation, as persistent requirements
… throughout the entire lifecycle.”

Software Testing for Radar & Electronic Warfare Systems 295

Choosing Appropriate Testing Infrastructure
There are several open source and commercial tools available to automate software testing. However,
because of the unique requirements of radar systems, there are some qualities and capabilities to consider
when assessing software test automation solutions. Keep the following in mind when evaluating solutions:

• Testing should cover the user journeys and interactions between the users and various devices in
radar systems, not just individual software components in isolation.
• There is often a diverse set of interfaces to be tested, from the command line to graphical user
interfaces (GUIs) like touch screens.
• As such, it is important to have the ability to test multiple endpoints built on entirely different
technologies and programming languages.
• Tests should be able to be conducted in a non-invasive manner, i.e., without access to source code.
• Testing solutions should exist on-premises and be operable without phoning home.

Testing requirements are not limited to individual systems, components, or pieces of software themselves,
but rather the supersystems of connected devices and software covering multiple endpoints. As such,
principles and practices from software test automation can be applied in the context of radar systems
and electronic warfare. The sheer diversity of devices and software means that any software test
automation solution must be sufficiently adaptable. This goes way beyond simple bug testing. Interactions
that real users can do must also be tested, such as using a touch screen, transferring data across a
network, or reaching into a database.

Creating a Testing Plan

Various types of testing should be included in your software testing plan:

• Regression testing — When certain components of the system are upgraded, will the previously
developed and tested software still work in the desired manner?
• Usability testing — Can operators of radar systems access and use the features needed to fully
take advantage of the systems’ capabilities?
• Performance testing — It’s important to replicate real-world conditions to ensure that the systems
can perform in the desired manner. Tests should look at peak conditions, operating capacity,
application and device integration, response time, and loading speed.
• Functionality testing — This covers testing system functionality, user workflows, and accessibility
considerations.
• Integration testing — Does the software interact with or draw information from other systems? If so,
these integrations need to be tested.
• Security testing — The software needs to stand up to the strict security standards and protocols
demanded by defense agencies.
• Exploratory testing — The testing solution should be able to proactively track down and identify
bugs. Software test automation can be augmented by artificially intelligent bug detection that will flag
up errors that human testers will miss.

Software Testing for Radar & Electronic Warfare Systems 296

Given the mission-critical nature of software testing, it’s important to build a coherent and robust software
testing plan. Consider the wider strategic goals of your radar systems, how users will interact with these
systems, and what human and technological resources you will need to carry out your testing plan.

AI as an Accelerator for Software Test Automation

Advances in artificial intelligence (AI) have direct consequences for testing the software of radar and
electronic warfare systems. AI can augment many areas of software testing: regression, usability,
performance, functional, integration, security, and exploratory.

In short, AI allows software interfaces to be interpreted and intelligently interacted within the absence of
direct human supervision or input. AI can then explore the various permutations of actions possible in the
interface, autonomously generating novel user journeys that may unearth otherwise unknown bugs and
defects. This exploration of an application or software system is called exploratory testing, a discipline
that has traditionally been heavily reliant on manual human testers. AI-augmented exploratory testing can
be a uniquely effective method to “identify system capabilities, limitations, and deficiencies,” a requirement
of DoDI 5000.89, ‘Test and Evaluation.’

Software test automation benefits from AI by first interpreting the text and visual elements on the
screen via optical character recognition (OCR) and computer vision, respectively. This requires the AI to
apprehend and interpret various screen elements, whether they are input controls (e.g. buttons or text
fields), navigational components (e.g. breadcrumbs or icons), informational components (e.g. notifications
or message boxes), containers, or other objects. Once a model of the application or system under test
has been constructed, intelligent agents (sometimes referred to as ‘bots’) will navigate the application and
perform actions.

Software Testing for Radar & Electronic Warfare Systems 297

The training set that allows the AI to intelligently navigate the software interface can be provided by two
general sources:

• Exploratory testing models learned from millions of user journeys across all types of software
and devices.
• User behavior observed from interaction with the specific application to be tested. The application
can be instrumented and event-level data from real usage is collected and used to train the AI.

AI is also helpful for the visual validation of software. Code or object-level testing tools can provide a
certain level of confidence that data and objects have loaded properly. However, certainty can only be
achieved if visual validation tools are used to ensure that the elements onscreen reflect the intended
outcomes. In other words, computer vision can be used to verify that the user interface is behaving
expectedly. When visual validation fails, this indicates to development teams that there is some behavioral
quirk in the software that needs to be addressed.

Intelligent agents can also be used to stress test systems for performance and load testing. Virtual users
can flood an application, with visual validation tools used to ascertain the impact on the user experience
of peak loads.

AI has a great deal to offer software teams in the radar and electronic warfare space. By augmenting
existing testing methods through AI and automation, developers can achieve a much greater
understanding of the true performance, behavior, and utility of their software in real-world scenarios.
Ultimately, this reduces the amount of ‘unknown unknowns’ for the modern warfighter.

SOFTWARE TEST AUTOMATION FOR AEROSPACE & DEFENSE

https://www.keysight.com/see/radar-ew-system-test

Software Testing for Radar & Electronic Warfare Systems 298

Chapter 27
New Age Requirements of Cyber Security
Connected networks such as air defense and air traffic control are vulnerable to cyber-attacks.
Search online and you will find examples of hacking and spoofing that disrupted defense networks
rendering them temporarily ineffective, allowing adversaries to gain tactical advantage. Not all
hackers have such perilous intent – their motivation may be espionage or simply penetrating a
secure system for “fun” as a test of skills and wits. With modern systems becoming increasingly
connected, and with regular software updates required for adaptive systems and machine learning,
securing your operations against cyber-attacks has never been more important.
Evaluating Security Effectiveness
One of the challenges of assessing security effectiveness is that it has been hard to quantify in the past.
Your security was either effective, meaning that you had not been breached, was not effective because
you had been breached, or worse, you thought it was effective simply because you didn’t know you had
been breached. Security teams have focused correctly on the prevention and detection of attackers.
Malware relies on network communication to download instructions and transmit sensitive data. Actively
blocking network probes, phishing clicks, and all traffic to and from untrusted countries dramatically
reduces your exposure. With those technologies and processes in place, it has become critical to obtain
a repeatable and ongoing measurement of your security posture.

SecOps Impact
Breach and attack simulation expose the gaps in your security environment, but it also helps you fix
the problems and then re-audit to confidently know your work has made your security stronger. For too
long, security professionals have often had to simply guess that what they did on a Monday helped the
company’s security on Friday. With breach and attack simulation, that is no longer a question. Now you
know the impact of your time and resources, allowing you to accurately measure security operations
effectiveness and costs.

Enabling Breach and Attack Simulation

Being aware of today’s threat landscape is vital to identify the weaknesses in your environment. Threat
intelligence comes in a variety of flavors and nearly every security product has a “threat intel” feed that
allows it to understand what to defend against. The end goal of threat intelligence is to provide
knowledge that allows you and your tools to make the right decision. Traditionally, threat intelligence has
been used proactively to either stop a known bad actor, perform incident response, or breach forensics.
Security teams are beginning to realize that the same knowledge about attack exploits and techniques
can be used proactively to harden security before an attack. To be proactive in using threat intel it is
critical to also have a long history of knowledge alongside the data, as well as expertise to understand
the latest threats.

A defense organization will be genuinely assured in its security posture only after having it tested. Being
“tested” is not a euphemism for suffering a breach. Being tested means deploying a breach and attack
simulation solution to continuously create security audits. Continuous breach simulation identifies weak
spots and misconfigurations to be more closely monitored. With breach and attack simulation, an
organization gains first-hand insight into how the attack will succeed — without opening the door to
their adversaries.

Additional Reading
Breach & Attack Simulation For Dummies, Keysight Technologies Special Edition

New Age Requirements of Cyber Security 300

Section 6

Gain Confidence, Achieve Realism

Chapter 28
Achieve Realism and Accuracy in EM Spectrum Operations
Using a scalable, high-performance commercial off-the-shelf (COTS) testing approach, you can create
threat simulations and analysis systems that keep pace with the modern electronic warfare environment.

The electromagnetic (EM) spectrum environment is evolving with increasing speed and complexity.
New threats emerge constantly, driving the need for electronic warfare (EW) systems that can
accurately identify and neutralize radar threats. Often, these responses include adaptive and cognitive
countermeasures. To provide the accuracy required for system performance, EW tests and evaluations
must adapt through significant advances.

This challenge faces increased pressure with high-fidelity complex emitters, as there is a need to simulate
and analyze high-density environments. As the EW threat environment continues to evolve, confidence
and reliability in EW system validation and verification requires modernization and improvements in the
test and evaluation process. To ensure consistency throughout the development, testing, and deployment
of EW systems, you must identify when or where an error occurred. With Keysight’s full range of COTS
building blocks, you can test and evaluate from digital modeling in research and development through
system integration, flight tests, and fully operational systems.

Automate Source Phase Coherent UXG Multi-port Multi-channel Signal

Calibration Agile Signal Sources EW System Analysis Platforms
Under Test

Achieve Realism and Accuracy in EM Spectrum Operations 302

Simulate EW Systems Under Test (SUT)
By performing digital modeling before hardware implementation of an EW system under test (SUT), you
reduce the risk of both development time and high program costs. Keysight’s SystemVue radar modeling
library helps to verify and analyze EW system processing, algorithms, and countermeasures by creating
digital models of the system and running simulations with environmental effects including multi-path
reflections, interference, jamming, targets, and clutter.

Using a COTS range of test and measurement equipment provides testing flexibility and adaptability
as the system moves from a digital model into hardware prototypes during hardware-in-the-loop (HIL)
testing. Choose from a range of arbitrary waveform generators and agile vector signal generators for
signal generation. You have a choice of oscilloscopes or digitizers for signal analysis. Using SystemVue as
a platform, you can control the hardware test setup.

• Analyze entire system including dynamic flight path, multi-emitters, jamming, and interferers
• Model environment, RF hardware, antenna, and phased array effects
• Reduce cost and time of field flight tests through simulation

SystemVue
TEST SCHEDULE CONTROL

INSTRUMENT CONFIGURATION

ADVANCE SIGNAL GENERATION

SUT CONFIGURATION

SUT OUTPUT CAPTURE

Signal SUT Signal

Generator Analyzer/Scope

REFERENCE RECEIVER

ADVANCED MEASUREMENTS

Achieve Realism and Accuracy in EM Spectrum Operations 303

Get Closer to Reality
With the rapid evolution in the EM spectrum environment, you need to quickly adapt to new, complex
threats on an ongoing basis. Scalable signal simulation enables you to create complex, high pulse density
EW scenarios and simulate Angle of Arrival (AoA) and kinematics (moving platforms) simultaneously.
Keysight’s capability centers on multiple coherent N5193A/94A UXG agile signal generators.

With an agile signal generator that can switch frequency and settle amplitude in the hundreds of
nanoseconds at different frequencies, you can accurately simulate radar threats and targets. When
looking for a solution to simulate your RF environment, make sure the product’s internal modulation
bandwidth is sufficient to cover your threat frequencies of interest. Our solution combines UXGs,
calibration hardware and software, and application software for pre-scripted or dynamic threat generation
scenarios. The application software has been configured to simulate an electronic battlefield with
thousands of emitters.

Pulse Descriptor Word

Based Operation Applies User RF Corrections
- 1Gb LAN or SSD streaming to > 1Mpps to Pulse Descriptor Word Data
- 10 GB LAN streaming to > 10Mpps - To system under test input port
Tx

Rx

Reports Pulse Counts EW System

- Received, played, expired, or collided
Under Test

The UXG agile signal generator’s ability to fast frequency hop with phase continuity and repeatability
makes it an ideal source to efficiently simulate complex threat environments across the signal generator’s
full 40 GHz range:

• Multiple pulse-Doppler radars at different frequencies while maintaining the original phase as the
signal generator hops from one emitter frequency to another
• EW scenarios with thousands of radar threat-emitters and millions of pulses per second with unique
antenna scans
• IQ custom complex modulation on pulse with the UXG Vector Adapter including linear and nonlinear
frequency modulated chirps over a 20 GHz range
• Scaling up the number of UXGs to increase pulse density while enabling pulse-on-pulse simulations
or multi-port angle of arrival (AoA) simulations
• AoA simulations with multiple UXGs and staggering identical pulses played out of different ports
(different UXGs) in time, phase, amplitude, or all three

Achieve Realism and Accuracy in EM Spectrum Operations 304

Timesaving approaches to PDW simulations
The UXG’s flexible architecture and legacy threat library import capability offer a replacement for current
RF sources or integration into new threat simulators. Easily use existing Pulse Descriptor Word (PDW)
libraries or create new ones using a variety of tools including Excel, Matlab, or Keysight’s N7660C
MultiEmitter Scenario Generation software or Z9500A Simulation View software1.

See, Capture, Analyze, and Understand Complex Environments

Analyzing and scoring EW stimulus and electronic attack resources presents unique challenges, as the
high-density environments feature wide-bandwidth and complex pulse modulation types. With Keysight’s
COTS analysis hardware, including signal analyzers, oscilloscopes, and digitizers, you can capture and
characterize the EW environment up to 110 GHz with wide modulation bandwidths.

The key is pulse analysis software. Using the Keysight N9067C and 89601B software, you can
differentiate threats with pulse-scoring filters based on characteristics such as pulse width, PRI, and
modulation type (including linear and non-linear modulation). You also can capture long scenarios with
efficient memory usage and make correlations and sidelobe measurements.

Dual-domain analysis with narrow bandwidth enables frequency domain analysis while wider bandwidth
targets time-domain analysis. Using the 89601B software, you can compare sequential pulses to assist
with radar output waveform validation with Pulse Similarity Scores. Pulse Train Searches ensure that radar
mode changes happen as expected.

Record, Score, and Analyze System Outputs

In the dynamic EM environment, EW systems must confront an onslaught of new signals. The radar
recorder can record and analyze pulsed signals in real-time, making it the ideal multi-channel system
to witness and verify the output of electronic attack systems. Using real-time measurement and PDW
scoring, Keysight’s radar recorder quickly verifies measured pulses. It increases confidence in testing
while significantly improving recorder scalability, capability, and support.

For example, the Z2099B family’s wideband, multi-channel design allows simultaneous two-channel
recording for analysis on both inputs and outputs. Real-time measurement and PDW scoring increase
confidence in testing while introducing the rapid verification of measured pulses. You can easily view
and analyze large amounts of PDW data via the analysis software. Staggered channel capture allows
for simultaneous recording and data offloading and analysis. This system family ranges from small and
transportable to fill systems, allowing the radar recorder to scale easily.

1 N7660C and Z9500A software are subject to US ITAR export regulations. For more information, contact your Keysight sales
representative.

Achieve Realism and Accuracy in EM Spectrum Operations 305

COTS EW Signal Generation and Analysis Building Blocks
Keysight provides commercially available building blocks to create EW threat simulation and analysis
systems. This includes solutions with:

• N5193A/94A Agile signal generators for multi-channel and multi-port threat simulations
• Arbitrary waveform generators for threat simulation for baseband or wideband verification
• Oscilloscopes, signal analyzers, and digitizers for wideband signal analysis
• Flexible FPGA tools and storage and streaming options for closed-loop simulations

Flexible EW Threat Simulation and Analysis Solutions

Keysight has a long history in measurement and calibration science. Work with our team of experts to
configure and design a scalable and flexible EW test system. Our solutions include full system integration,
automated multi-source, and system-level calibration.

• Create high-density, AoA simulations with flexible multi-port configurations

• Ensure coherence across multiple sources with calibration of amplitude, phase, and time
• Threat simulators compatible with MESG software or other dynamic PDW-based scenario generation
systems
• Capture the signal environment using a multi-channel, wide-bandwidth streaming recording of RF and
processed signals
• The integrated combination of hardware, firmware, and software performs signal selection,
downconversion, digitization, signal processing, and data storage
• Apply post-capture analysis software to RF test and signal-processed recordings

Real-Time Sequenced PDW Files

Often, real-time sequenced streaming is necessary for EW signal generation. For example, such capability
is useful during an over-the-air test or open range test to simulate multiple ground-based radars. The
UXG signal generator synchronizes to any time module that can output PPS - for example, GPS. With
multiple boxes synchronized, they are triggered at the same universal time clock. As a result, the operator
can do the following:

• Control several UXG stacks located over a significant distance from one central location
• Use the same PDW library and threat simulation files from early prototype and system integration
testing to verify in-flight/operation receiver & processor effectiveness and stability and
• Play the same files and simulations at the corresponding UTC across different labs or locations

Achieve Realism and Accuracy in EM Spectrum Operations 306

Pre-Mission Go/No-Go Testing
Once a system is deployed into the field, there may be a need to perform pre-mission tests to verify
operational readiness. Keysight’s FieldFox handheld microwave analyzers offer benchtop performance out
in the field across multiple terrains and extreme environments including clean room to desert, sea, tropics,
and arctic. Gain confidence in mission-critical measurements including:

• Receiver Test
- noise figure
- functional test with CW source
• Emitter Verification
- verify signal output and characteristics with a power meter, spectrum analyzer, and real-time
spectrum analyzer to 50 GHz
- pulse profiling to 40 GHz with USB peak power sensor
• Op check entire RF chain or individual components, radiated or closed-loop
- antennas, cables, converters, amplifiers
- distance-to-fault and Time domain reflectometry (TDR)
• Op Check GPS
- evaluate carrier-to-noise density (C/N) and distribution amplifiers

Achieve Realism and Accuracy in EM Spectrum Operations 307

Support
Keysight offers a broad portfolio of services and support to assist engineers working on electronic warfare
programs. We understand that engineers count on accurate, repeatable measurements to ensure mission
success while meeting budget and schedule requirements. Inaccurate measurements and system
downtime affect yield and the risk of a device failing during operation. The emergence of new and unique
threats drives a need for constant modernization.

To address this challenge, you can:

• Implement an optimal migration strategy with Technology Refresh Services to modernize test
equipment to the latest technology as soon as it is available with upgrade and Trade-In Services
• Have confidence that your instruments are performing to specification by utilizing Keysight’s global
network of service centers across multiple countries
• Avoid the need to disassemble and reassemble test systems and improve true yields with System
Calibration Services

One of the long-standing trends in the industry is the need for long-term support for sometimes multi-
decade programs with little to no budget for upgrades:

• Use Extended Support plans to ensure the availability of parts for repair and calibration procedures
for legacy instruments
• Take advantage of Keysight’s expertise in test and sustainment planning through Consulting Services
• Leverage One-Stop Calibration Services to reduce logistical complexity and lower costs with one
point of contact for the calibration of all your test assets

As the demand for better performance and newer technologies continually drives more complex designs,
with narrow test margins and longer test times. The complex test systems composed of instruments from
multiple vendors must be managed. To help mitigate these challenges, you can:

• Dramatically reduce test times and improve test system efficiency through Keysight Process Analysis
Services
• Improve operational performance and ensure ongoing accuracy through accredited and standards
lab calibration on Keysight and non-Keysight electronic instruments
• Manage downtime with loaner services, onsite calibration, and onsite resident professionals
• Minimize risk and reduce costs by leveraging System Calibration Services to ensure that your test
systems are performing to the test system uncertainty we calculate

Achieve Realism and Accuracy in EM Spectrum Operations 308

Helping You Prevail in the EM Environment
With the accelerating evolution of EW threats, dramatic improvements in tests and evaluation are not
optional. Welcome to a new era: Keysight is stepping forward as a commercial collaborator, creating and
delivering the rapidly adaptable solutions you need to succeed far into the future. Our mission is to work
with your team to ensure enhanced realism and greater confidence.

Engage with Keysight — solutions and services — and take your lab to the next level.

Note: The EW software is subject to ITAR export regulations. For more information, a live demonstration, or a trial
license, please contact your Keysight sales representative.
Appendix
Acronyms

AAI Air-to-Air Intercept DAC Digital to Analog Converter

ACPR Adjacent Channel Power Ratio DANL Displayed Average Noise Level

ADC Analog-To-Digital Converter DBF Digital Beam Forming

ADS Advanced Design System DDS Direct Digital Synthesis

ADT Automatic Detection and Tracking DOA Direction of Arrival

AESA Active Electronically Scanned Array DRFM Digital Radio Frequency Memory
AEW Airborne Early Warning DTM Digital Terrain Model

AFC Automatic Frequency Control DUT Device Under Test

AFR Automatic Fixture Removal EA Electromagnetic Attack

AGC Automatic Gain Control ECC Error Correcting Code

AI Artificial Intelligence ECCM Electronic Counter-Countermeasures

AM Amplitude Modulation ECM Electronic Counter Measure

AoA Angle of Arrival EDA Electronic Design Automation

APAR Active Phase Array Radar EIRP Effective Isotropic Radiated power

AUT Amplifier Under Test ELINT Electronic Intelligence

AWG Arbitrary Waveform Generator EMS Electromagnetic Spectrum

AWGN Additive White Gaussian Noise EMV Electromagnetic Vulnerability

BITE Built in Test Equipment EOB Electronic Order of Battle

BSD Blind Spot Detection EPM Electronic Protective Measures

BW Bandwidth ESA Electronically Steerable Array

CAD Computer Aided Design ESM Electronic (Warfare) Support Measures

CEMA Cyber and Electromagnetic Activities ETP Equivalent Transmitter Power

CFAR Constant False Alarm Rate EVM Error Vector Magnitude

COHO Coherent Oscillator EW Electronic Warfare

COTS Commercial off the Shelf FAR False Alarm Rate

CW Continuous Wave FCA Frequency Converter Application

Acronyms 311
FCC Fault Collection Unit MMIC Microwave Monolithic Integrated Circuit
FCW Forward Collision Warning MSSR Monopulse Secondary Surveillance Radar
FDOA Frequency Difference of Arrival MTD Moving Target Detection

FFT Fast Fourier Transform MTI Moving Target Indication

FM Frequency Modulation MTTR Multi Target Tracking Radar

FMCW Frequency Modulated Continuous Wave NFA Noise Figure Analyzer

FOM Frequency Offset Mode NIST National Institute of Standards and Technology

FPGA Field Programmable Gate Array NVNA Non-Linear Vector Network Analyzer

FSK Frequency Shift Keying OCR Optical Character Recognition

GCA Ground-Controlled Approach OTH Over-The-Horizon

GPR Ground Penetrating Radar PA Power Amplifier

GTC Gain Time Control PAR Phased-Array-Radar

HIL Hardware in Loop PAR Precision Approach Radar

HPA High Power Amplifier PDW Pulse Descriptor Word

IF Intermediate Frequency PDF Probability Distribution Function

IFF Identification Friend or Foe PESA Passive Electronically Scanned Array

IMD Inter Modulation Distortion PLL Phase Locked Loop

ISL Integrated sidelobe level PNTS Phase Noise Test System

LNA Low Noise Amplifier PRF Pulse Repetition Frequency

LO Local Oscillator PRI Pulse Repetition Interval

LPF Low Pass Filter PRN Pseudo-Random-Noise

LPI Low Probability of Intercept PRT Pulse Repetition Time

LRU Line-Replaceable Unit PSD Power Spectral Density

LVDS Low Voltage Differential Signaling PSL Peak sidelobe level

MBE Model Based Engineering PW Pulse Width

MDF Mission Data File RADAR Radio Detection and Ranging

MFR Multi-Function Radar RBW Resolution Bandwidth

MIMO Multiple Input Multiple Output RCS Radar Cross-Section

Acronyms 312
RDF Range and Direction Finding STC Sensitivity Time Control

RF Radio Frequency SUT System Under Test

RS Ramp Slope T/R Transmit/Receive

RWR Radar Warning Receiver TBP Time-Bandwidth Product

RX Receiver TDOA Time Difference of Arrival

SAM Surface-to-Air Missile TDR Time Domain Reflectometer

SAR Synthetic Aperture Radar TMA Terminal Maneuvering Area

SDR Software Defined Radio TOI Third Order Intercept

SFDR Spurious Free Dynamic Range TRL Thru, Reflect, Line(calibration)

SIF Selective Identification Feature TRM Transmitter-Receiver Module

SMC Scalar Mixer Calibration TWT Traveling Wave Tube

SMR Surface Movement Radar TX Transmitter

SNR Signal-to-Noise Ratio ULA Uniform Linear Array

SOLT Short, Open, Load, Thru (calibration) UUT Unit Under Test

SPI Serial Peripheral Interface UWB Ultra-Wideband

SSA Signal Source Analyzer VCO Voltage Controlled Oscillator

SSB Single Side Band VMC Vector Mixer Calibration

SSPA Solid-State Power-Amplifier VNA Vector Network Analyzer

SSR Secondary Surveillance Radar VSA Vector Signal Analyzer

STALO Stable Local Oscillator VSG Vector Signal Generator

STAP Space-Time Adaptive Processing VSWR Voltage Standing Wave Ratio

(The following acknowledgement applies to the acronyms section only)

Publisher: Christian Wolff (Revised by Karina Hoel and Keysight Technologies)

Text is available under the GNU Free Documentation License (https://www.gnu.org/licenses/fdl-1.3.html), and the
Creative Commons Attribution-Share Alike 3.0 Unported (CC BY-SA 3.0) (https://creativecommons.org/licenses/
by-sa/3.0/)license, additional terms may apply.

Acronyms 313
Additional Resources
Keysight Technologies Electronic Warfare Test and Evaluation Solutions, 5992-3476EN

Multichannel Coherent Electronic Warfare Analysis Solution, 3121-1019EN

Electronic Warfare Threat Simulation Solutions Catalog, 7120-1247EN

PathWave Design Software, 5992-4019EN

PathWave System Design, 3121-1074EN

Electronic Warfare Signal Generation: Technologies and Methods, 5992-0094EN

N5193A UXG Agile Signal Generator Data Sheet, 5992-0092EN

N5194A UXG Agile Vector Adapter Data Sheet, 5992-2228EN

Signal Studio for Multi-emitter Scenario Generation, 5992-0405EN

Creating Multi-Emitter Scenarios for Radar and Electronic Warfare (EW) Testing, 5992-2902EN

Z9500A Keysight Threat Simulation Overview, 5992-4339EN

N9040B UXA Signal Analyzer Data Sheet, 5992-0090EN

N9067C Pulse Analysis Application Technical Overview, 5992-1384EN

Pulse Analysis – PathWave Vector Analysis (89600 VSA) Technical Overview, 5992-4197EN

Radar Measurements, 5989-7575EN

Measuring Radar Signals with Vector Signal Analyzers and Wideband Instruments, 5992-1580EN

PNA and PNA-X Series Microwave Network Analyzers, 5990-4592EN

Field Fox Handheld Analyzers Technical Overview, 5992-0772EN

Techniques for Precision Validation of Radar System Performance in the Field, 5991-4107EN

Keysight Care, 5992-3373EN

314
This information is subject to change without notice. © Keysight Technologies, 2021, Published in USA, December 1, 2021, 7121-1111.EN