Text preview for : 5990-7532EN Time Sidelobe Measurements to see Performance of Compressed-Pulse Radars - Application N part of Agilent 5990-7532EN Time Sidelobe Measurements to see Performance of Compressed-Pulse Radars - Application N Agilent 5990-7532EN Time Sidelobe Measurements to see Performance of Compressed-Pulse Radars - Application Note c20140725 [7].pdf
Back to : 5990-7532EN Time Sidelobe | Home
Keysight Technologies
Solutions for Using Time Sidelobe Measurements to
Assess the Performance of Compressed-Pulse Radars
Application Note
Radar Measurement Series
Overview the effect of such a reflection when evaluating the overall functional
performance of a frequency chirped, pulse compression radar.
In a radar system, the use of modulation on pulse for "compression"
provides enhanced spatial resolution as well as extended range for a Pulse compression radar
given output power level. Consequently, this technique is widely used Waveform
exciter
DAC PA
Transmitter
in current- and next-generation radar systems. Antenna
STALO
COHO
Unfortunately, traditional RF pulse measurements become less
effective predictors of performance in radars that use pulse I Synchronous
IF LNA
Q I/Q detector
compression. For example, the width of an uncompressed radar Receiver
pulse is directly related to spatial resolution. In contrast, the
Mismatch
Mismatch
resolution depends on pulse width, chirp bandwidth and chirp
linearity in a compressed radar system that uses linear frequency
modulated (LFM) chirp pulses.
Within the field of radar development, a technique called the time Reflection Reflection
sidelobe level (SSL) measurement has emerged as a viable solution Internally generated echo
to predict performance. This method distills a wide range of potential
signal impairments down to a simple metric that can be used to
determine if radar performance will fit the intended application. Figure 1. Mismatch reflections may cause ghost echo returns
This application brief defines key problems related to measurements
Other problems from within the radar may involve other component
of compressed-pulse radars, describes the time sidelobe method, and
impairments that can affect the results at the demodulator.
outlines the practical uses of this approach.
Examples include unintended phase noise, amplitude modulation,
Problem reflections and group delay. These routinely distort what is actually
detected in the receiver, limiting the dynamic range and detection
As suggested above, the modulation parameters used in compressed accuracy of the radar.
radars clearly affect system performance. One of the overall problems
is in the determination of "what's good enough?" with regard to the
attributes of components used to build radar subsystems. Similarly,
in a diagnostic scenario, it is difficult to assess overall performance
without separately measuring the traits of individual components.
As an example, it can be difficult to judge the affect of component
performance on any impairment added to a modulated pulse--and
errors in these measurements routinely result in costly over-specifying
of components. Consider a frequency-dependent mismatch reflection
of an IF filter used in a radar receiver (Figure 1). Typically, lower-
frequency IF filters have significant signal delays. When these
combine with internal reflections from a desired radar pulse the result
may be a "ghost" echo return. It might be very difficult to determine
Solution Windowing functions
Amplitude weighting of the output signals is generally used to
One way to meet these challenges is to distill the characterization reduce the time sidelobes to an acceptable level. As a side effect
of a wide range of potential signal impairments down to a simple the signal weighting will result in the loss of signal to noise ratio.
metric, preferably one that can be used to determine if radar Some of the more commonly used windowing functions are shown
performance will fit the intended application. Time sidelobe
in table 1 with their suppression levels and signal to noise losses.
measurements provide an effective way to use known-good
test equipment--with internal impairments calibrated out of the Table 1. Popular windowing functions and their effects
measurements--and mathematically consistent processing of
measurement data to accurately characterize the performance of a Weighting Peak Sidelobe S/N Loss (dB)
compressed-pulse radar. Function Level (dB)
Uniform -13.2 0
What are time sidelobes? Hamming -42.8 1.34
Sometimes referred to as range sidelobes, time sidelobes are a Hann -32 1.4
result of using pulse compression techniques. They are produced Blackman -58 2.37
when the ideal radar return is convolved with the response of the
Blackman-Harris (3 term) -67 2.33
less than ideal correlation filter during the compression process
or when a non-deal radar return is convolved with the response
of the less than ideal correlation filter or some combination of Figure 2 graphically shows the characteristics of the Hamming
the two. This causes some of the energy in the return pulse to lie function.
outside the pulse bandwidth. In the time domain this is indicated
by a spreading in range (time) of the return pulse, particularly in
the presence of ground clutter or transmit signal anomalies caused
by imperfections in the transmitter path.
Since the correlation filter in modern radar is nearly always
implemented digitally within a DSP rather than with an analog
Standing Acoustic Wave (SAW) filter, the resulting compressed
pulse waveform is mathematically deterministic and repeatable
and as such, easily optimized through simulation.
Figure 2. Hamming windowing function and associated frequency response
2
F ( t ) = A cos( wc t + 1 ut 2 )
2
Time side lobe level
(SLL) is a quality metric
Pulse Compressed
width pulse width
SLL
Correlation
filter
Linear FM chirp pulse Compressed pulse
Figure 3. Pulse compression and the Time Side Lobe Measurement
Applying the time sidelobe method SLL measurements made. The pulse generation platform should
provide repeatability when creating and recreating the pulses
The detector in a radar receiver correlates the transmitted signal
for the measurement stimulus and for the relative comparison
with echoes and noise received over time. When the received
measurement.
signal matches the sent signal a correlation peak occurs and
target detection is marked at that time.1 Alternatively, the ideal compressed pulse could simply be created
mathematically and compared to the measured pulse without
An ideal correlation peak would be infinitely narrow, have a value ever physically generating it. The limitation here is that the relative
of one, and be surrounded by noise-like sidelobe levels. At the comparison is always with the ideal which may not always be
other extreme, the correlation between a pure signal and pure desired. At times there may be a need to measure SSL between
noise is zero. Figure 3 indicates the process of pulse compression two points within the system in trying to track down the specific
and the time sidelobe measurement. source of an anomaly.
System impairments ranging from imperfectly generated
compressed pulses to internal reflections from filters can cause Three tools suggested for creating the pulsed waveform may use
different methods but the end result is the same: a combined or
correlation sidelobe levels well above the noise floor (Figure 4). separate I and Q waveform file(s) that can be directly downloaded into
Because it is difficult to judge the effect of such impairments on the waveform memory of an Arbitrary Waveform Generator (AWG).
a compressed pulse, using the time sidelobe level technique
provides quantitative measurements of transmitted pulse shapes SystemVue: A system modeling tool, SystemVue with the optional radarmodel
library provides signal processing reference models for exploring trade-offs in
and received signals. radar system architectures for Pulsed Doppler, FMCW, Digital Array, and UWB
Radars. It enables scenario modeling by adding targets, clutter, fading, noise,
An accurate measurement of SLL requires a filter that is perfectly interferers, and the RF effects necessary for realistic system analysis and early
correlated to the desired pulse shape. The first step is to build an R&D verification using connections to live test equipment.
www.keysight.com/find/SystemVue
ideal waveform that represents the desired compressed pulse:
bandwidth, pulse width, and chirp or modulation characteristics
Signal Studio for pulse building: Enables flexible generation of complex,
are essential parameters. Modeling can be performed with wideband pulse patterns using the E8267D PSG or E4438C ESG vector signal
software such as SystemVue or Signal Studio for pulse building generators. Custom pulse shaping, modulation, antenna patterns, and user-
from Keysight Technologies, Inc. or MATLAB from The MathWorks. defined pulse patterns are easily achieved with the straightforward graphical user
interface or with your own test executive using the COM-based API. Add the
The mathematically generated, ideal (i.e., repeatable with no N603xA/N824xA/M933xA wideband AWG to Signal Studio for pulse building
impairments added) representation of the compressed pulse can such as signal processing, signal modulation, digital filtering, and curve fitting.
be stored in memory and recalled later to correlate measured www.keysight.com/find/signalstudio
waveforms and enable calculations of SLL.
MATLAB: A software environment and programming language created by
MathWorks and now available directly from Keysight as an option with most sig-
The FM chirp pulse waveform for the SLL measurement and nal generators and signal analyzers. MATLAB extends the capabilities of Keysight
calculation should be designed to imitate the operational signal analyzers and generators to make custom measurements, analyze and visu-
waveform of the radar system. For systems that feature multiple alize data, create arbitrary waveforms, control instruments, and build test systems.
It provides interactive tools and command-line functions for data analysis tasks
operating modes multiple waveforms should be used and multiple such as signal processing, signal modulation, digital filtering, and curve fitting.
www.keysight.com/find/MATLAB
1. As a reminder, the time differential between "send" and "receive" is
related to the distance between the radar and the target.
3
When measuring pulse waveforms that will be correlated for SLL Once these preparations have been complete, the measurement
calculations, instrument calibration and waveform correction are can be performed. Within the VSA software, the measured
other important factors. The reason: The wideband nature of most frequency data (real and imaginary) is multiplied by the ideal
compressed pulses gives rise to potential inaccuracies in both pulse. This result is processed with the inverse fast Fourier
phase and amplitude versus frequency within the measurement transform (IFFT) function to produce the time cross correlation
instrument. To prevent these inaccuracies from affecting SLL needed for the SLL measurement.
measurements, a measuring receiver with built-in equalization
and signal generation software or hardware with predistortion Measured (t) Ideal (t) = IFFT [Measured(f) * conj[Ideal(f)]]
capabilities is essential. Where
Measured (f) = window * FFT (Measured (t))
Making time sidelobe measurements Ideal (f) = window * FFT (Ideal (t))
The measurement process starts with detailed knowledge of the
ideal compressed pulse, as described above. The next requirement
is a suitable broadband signal analyzer, oscilloscope or logic
analyzer with vector signal analysis (VSA) software. Example
instruments include Keysight's PXA and MXA signal analyzers,
Infiniium 90000X-Series oscilloscopes and 16900 series logic
analyzers. These all support the Keysight 89600 VSA software,
which supports more than 70 signal formats and is capable of
implementing the mathematics needed to make and display the
time sidelobe measurement. The VSA software can run on a PC
or inside instruments such as those mentioned here.
The instrument is used to acquire and digitize the measured
waveform. The VSA software can be configured to use the