=Paper=
{{Paper
|id=Vol-2005/paper-03
|storemode=property
|title=Radar Simulator's Signal Processing in the Distance Range down to the Zero Value
|pdfUrl=https://ceur-ws.org/Vol-2005/paper-03.pdf
|volume=Vol-2005
|authors=Alexander S. Bokov,Vladimir G. Vazhenin,Sergey A. Margilevsky
}}
==Radar Simulator's Signal Processing in the Distance Range down to the Zero Value==
https://ceur-ws.org/Vol-2005/paper-03.pdf
Radar Simulator’s Signal Processing in the
Distance Range down to the Zero Value
Alexander S. Bokov, Vladimir G. Vazhenin, and Sergey A. Margilevsky
Ural Federal University, Yekaterinburg, Russia,
a.s.bokov@urfu.ru
Abstract. The paper considers devices designed to simulate a time-
frequency structure of a radar signal reflected from a target or underlying
surface. Proposed simulator can be used to simulate false radar targets
including those located closer than the minimal delay the devices may
implement in conventional simulators. An effective way is suggested to
reduce the simulated range. The method works by real time application
of an additional frequency shift. It may be useful to simulate the com-
bat operation of a radar system, as well as to simulate echo signals of
radio altimeters when probing by signals with various types of linear fre-
quency modulation. The dynamic modification of simulating parameters
is carried out in accordance with parameters of the input emitted signal.
Keywords: Altimeter, radar target, radar echo simulator, linear fre-
quency modulation, digital signal processing
1 Introduction
A device for simulating a false radar target is intended to simulate the time-
frequency structure of a radar signal reflected from an underlying surface or
from one or more targets located in a fixed direction. It can be used to simulate
the following echoes from false target including those located closer to the carrier,
and from underlying surfaces for radio altimeters (flight altitude meters).
Simulators may be useful to verify and even try to deceive the combat ope-
ration of radar systems in different circumstances [1, 5].
Here, radar systems, which operate with linear frequency modulated continu-
ous waves (FMCW) are discussed, because they are widely used in manned and
unmanned aircrafts and helicopters especially at low distances and altitudes.
Their signals are convenient to effective processing with low level of emitted
power [6, 12].
In order to qualitatively simulate the frequency-time structure of the reflected
radar signals, we can use methods to form the reflected signal as a sum of signals
reflected by number of different small flat facets (areas) of the surface; such facets
are sufficiently smaller in comparison with the irradiated area. Facets may be
equivalent to bright points in some target and surface models. Such approach is
applicable and correct if their reflection coefficients are related to their physical
areas, orientations, and local surface “brightnesses”, i.e., the radar cross section
corresponds to the carrier wavelengths.
22
2 Typical simulator implementation with a delay line
A typical device for simulating radar targets with high resolution [1, 5] is shown
in Fig. 1. The emitted pulse from the radar, for which a radar portrait is crea-
ted, passes through the receiving (intercept) antenna, amplifier, bulk coarse de-
lay unit, accurate delay unit, set of modulators, and adder to the output of
the simulator. The bulk delay unit performs a time delay corresponding to the
distance to the nearest bright point of a simulated target. A delay line with
taps provides simulation of set of the bright target points. Amplitude and phase
modulations are performed by using external reference signals Ui , which corres-
ponds to the targets characteristics. From the output of the modulator, signals
simulating the corresponding bright points are directed to the adder and then
to the transmitting (retransmitting) simulator antenna.
Fig. 1. Scheme of the high resolution simulator based on a tapped delay line
There exist many similar devices. The described architecture corresponds to
the device for simulating real-world radar portraits [1] because of its structure
and operation principle; see, for example, the “Radio frequency signature aug-
mentation system” [2], the “Electromagnetic target generator” [3], the “Method
for deceiving a sonar or radar detector, and a decoy for implementing the
method” [4], and the “Method of augmentation of radar techniques” [5].
In practical application of the described methods and devices to simulating
radar portraits with variable modulation parameters, the significant problem
arises in simulating targets with ranges shorter than the distance to the carrier
(or vehicle) that needs to be protected (hided) from the operating radar.
The same difficulties appear when we use signal simulators to study the
characteristics of radio altimeters in seminatural modeling of work in a labora-
tory. It is impossible to simulate a signal with a signal propagation delay smaller
than the intrinsic delay caused by transition time in processing channels and
waveguides.
Now, the minimum delay is 40 nsec even in the best known schemes of digital
signal memory (DSM), which corresponds to the range of 6 m. With the use
of amplifiers, filters, attenuators, and connecting cables in real simulators, the
corresponding minimum simulated range is 14–60 m and more. It limits both the
ability to hide the real position of the carrier protected from the high-resolution
radar and the ability to simulate low altitudes when testing radio altimeters.
23
It is known that both the Doppler shift and the time delay of the signal, which
has a linear frequency modulation (FM) or an asymmetric sawtooth modulation,
can be simulated by a corresponding offset of its carrier frequency [10, 11].
Therefore, to reduce the minimum simulated height and compensate its own
delay in every hardware implementation, it is possible to use a certain frequency
shift. For example, an FMCW altimeter with an up-chirp sawtooth modulation
will register an equivalent low altitude if during the signal generation the addi-
tional frequency shift ∆f is applied to the direction, which is approaching the
transmitted ftx (t) and received frx (t) signals in the analysis window (measuring
part of the signal inside each repetition interval). See Fig. 2 where the frx (t)
graph is upshifted in the direction to be close to the ftx (t) graph.
Fig. 2. Simulation of an equivalent lower altitude for up-chirp FMCW altimeters
Frequency of the heterodyning signal (so-called beat signal) at the output
of the mixer in FMCW device is the difference between the frequencies of the
following two signals:
f (t) = |ftx (t) − frx (t)| ' const = F b (1)
where F b is the value of the beat frequency of the signal at the altimeter mix-
ers output inside the analysis window excluding the transition windows (the
beginning-and-ending periodical areas of modulation frequency jumps in ftx (t)
and frx (t) graphs).
It is seen (Fig. 2) that the positive frequency shift ∆f for the signal delayed
by some minimum delay τmin will lead to a decrease in the average beat frequency
F b and, as a result, to a decrease in the measured altitude or distance.
The first disadvantage of this method is the dependence of the required sign
of the frequency shift ∆f on the direction (type) of the linear ramp fashion slope
in up-chirp or down-chirp of the FMCW radar.
The second disadvantage is since the triangular wave modulation in the
FMCW, which is the most widely used scheme [12]. Consider the case when
we use only the positive frequency offset for the signal frx1 (t) signal (delayed by
τmin and shifted upward by ∆f relative to the signal ftx (t)) in the both half-
periods of the analysis windows (excluding the transition windows). Here, two
values of the beat frequencies are formed: the F bmin value in one half-period and
the F bmax value in the second half-period as by the frx1 (t) graph in Fig. 3
24
Fig. 3. Simulation of the lower altitudes in FMCW altimeters with the triangular
modulation
If the measurement unit of such radar operates on the leading edge of the
spectrum, then the problem of reducing the measured range could be solved.
But, let in the example of a typical FMCW device a radar system uses a range
measurement unit for the beat frequency corresponding to the “center of gravity”
of the spectrum averaged over the whole repetition interval [12]. Then such
a bifurcation of the original spectrum harmonics will not affect desirably the
measured distance in the radar system.
The simple replacement of “+∆f ” by “–∆f ” will only lead to a change in
the sequence of half-cycles with F bmin and F bmax , see the frx2 (t) in Fig. 3.
3 Sum of two frequency-shifted copies of the emitted
signal
For mentioned cases, it is suggested to generate a reflected signal as the sum of
two frequency-shifted copies of the emitted signal [7], see Fig. 4. It is effective for
FMCW and long pulse signals with linear FM when the measured frequency shift
is used to evaluate and accurately specify of the distance to the radar target.
Fig. 4. Simulation of lower altitudes for triangular modulation FMCW altimeters
25
The frequency offset is performed of the same value ∆f , but with opposite
signs: “+∆f ” and “–∆f ”. Here, ∆f is the parameter selected approximately
equal to or larger than the selective locking/tracking filter in the radar receiver.
The simulator comprises a mixer and a delay forming unit. The fref is the
frequency formed at the output of the mixer. With its help, the value of the
rate of change of the modulation frequency (Vf = 2W/TM , where W is the
modulation bandwidth and TM is the repetition interval, Fig. 3) can be evaluated
in the delay forming device.
Further, a delay value τ for the adjustable delay line is calculated by using
also the desired offset of the delay “∆τ ” coming from the external device
τref
τ = ∆τ + ∆f /Vf − τmin = ∆τ + ∆f − τmin , (2)
fref
where τmin is the own (internal) delay in the simulator circuits; Dτ is the required
offset delay with decreasing the simulated range or compensating for its own
delay (when the value has the minus sign); τref is the value of the signal’s taken
delay from any convenient output of the multiple-tapped delay line (see Fig. 4).
So, the original signal in the adjustable delay line is delayed by the delay τ .
In this case, the delay value τ determines the displacement of the entire portrait
of the simulated target in the range to a smaller one (for τ < ∆f /Vf ) or higher
(for τ > ∆f /Vf ). For example, when τ =0, the distance to the simulated target
will be reduced by
c τmin c ∆f
∆Rmax = = , (3)
2 2 Vf
where c is the speed of light.
The value of the parameter ∆f is chosen approximately equal to or greater
than the bandwidth of the selective capture and tracking filter in the radar
(altitude in the altimeter). However, in compliance with the condition for the
correct processing the received signal in the radar, ∆τ0 TM and, therefore,
∆f TM Vf .
Thus by [7], the device in Fig. 4 allows simulating two identical targets re-
gardless of the direction and signs combination of the rate of linear frequency
change. In this case, the first target (main target) can be simulated at a range
shorter than the range of the radar carrier. But the second target will be further
in range and, with the appropriate choice of parameters, will not interfere with
the correct tracking of the main target. The result achieved is the simulation of
a target with a range greater or less than the range of the carrier both for analog
and digital signal processing.
4 Commutation instead of summation
It is possible to modify the structure in order to improve the performance and
simplify the hardware implementation of the radar portraits simulator [7] of Fig.
4 when probing by such signals with various types of the linear FM. To simplify
26
the hardware implementation, instead of the second adder and the pair of fre-
quency shifters, the only one controlled frequency shifter can be used. Moreover,
only one sign of the frequency shift value ∆f is sufficient for the simulator to
work with both the triangular and asymmetric sawtooth modulated signals.
An example of decreasing the simulated range for the triangular FMCW is
shown in Fig. 5. Here, we use the periodically changing the plus sign or the minus
sign for a given frequency shift value ∆f . So, the entire signal of the previously
generated portrait of the simulated target takes the offset in its distance.
Fig. 5. Range harmonics and a beat signal spectrum while simulating the lower distance
Periodic change of the sign is carried out on regular intervals, and we denote
this value ∆t. Then, with continuous emission, the output signal will contain
∆t length segments of the emitted signal with phase jumps at the instant of
the sign change ±∆f . In the frequency domain, this will result in harmonics
corresponding to the sum, frequency differences of the “useful” signal, and to the
sign changing frequency Fsign multiplied by an integer [8], where Fsign = 1/(2∆t).
Let we select the Fsign (frequency of changing the sign) several times lower
than the average carrier frequency f0 , but above the band of “useful” signal
modulation frequencies (for example, at ∆t TM ) and take into account the
27
actual presence of limiting frequency filters in all radar receivers. Under this
the resulting signal in the working region will be equivalent by the spectral
composition to the signal formed by the conventional summation of the signals,
i.e., as in Fig. 4.
The principle of replacement of signal summation by signal “commutation”
was proposed and implemented in target simulators when probing with suffi-
ciently long radiated signals [8]. In the case of FM signals, it is also true, because
the spectral density of the radar probing signals with a chirp at each instant is
concentrated in a narrow frequency band. So, if parameters of the simulator are
chosen correctly, the interruption of the instant spectrum and the appearance
of extra phase jumps will not affect the operation of typical radar. Under this,
the target searching, locking, and tracking are performed without taking into
account the phases of the signals with averaging into several resolution elements
and, as a rule, in several repetition and scanning intervals.
Therefore, as a result of replacing the adder with the switchable modulator
as in Fig. 4, we get a bifurcation of the spectra at each instant and in the first
and second half-periods. Two pairs of spectral envelopes (for the point target
those are only harmonics F bmin and F bmax ) in the general spectrum of S(f ) will
be spaced along the frequency axis by 2∆f , see Fig. 5. Thus, for the case of a
FMCW radar with an arbitrary chirp slope, ∆f is greater than the bandwidth
of the bandpass (lowpass) beat signal filter. So, the high-frequency harmonics
will be suppressed or discarded, because the “second simulated range” are far
from the simulated target (surface). Moreover, and the real measured range value
will correspond to the “center of gravity” of the low-frequency envelope of the
spectrum with a smaller value of the “first simulated range” (altitude).
In Figure 5, the harmonic f τmin (t) are also shown for comparison. It corres-
ponds to a signal with real minimum delay τmin with none of frequency shift.
Its position in both half-periods of the modulation period is constant. So, it hits
(Fig. 5) the tail part of the low-frequency envelope of the spectrum of the beat
signal, i.e., to the right of the harmonic F bmin obtained by the proposed design
of the simulator.
Thus, we simplify the hardware implementation of the simulator in Fig. 4 in
the following way. Instead of the second adder and two frequency shifters, the
sign switcher, one frequency shifter, and, also, additional adjustable amplifier
are introduced. The device is shown in Fig. 6.
Fig. 6. Simulator with the switcher of the sign of the frequency shift
28
The simulator in Fig. 6 receives an emitted pulse from the radar, for which a
radar portrait is created, by the receiving antenna through. Further there are the
amplifier, multi-tapped delay line, set of modulators, adder, adjustable delay line,
adjustable amplifier, and frequency shifter to the output of the simulator. The
multi-tapped delay line provides simulation of bright points of the target(s) with
individual delays. Individual amplitude and phase modulations are performed
and they use the appropriate coefficients generated by the external device.
The output signal is obtained by shifting the frequency of the signal of the
generated radar portrait of the target, and the frequency shift is performed
alternately by the “+∆f ” and “–∆f ” units.
To implement this, the sign switcher is used. It inverts the frequency offset
value through time-equal intervals ∆t, i.e., frequency of signal is changed at the
output of the sign switcher Fsign = 1/(2∆t). Let the value of Fsign be selected
several times lower than the average carrier frequency f0 and outside the band
of the “useful” modulation frequencies of the radar signal; and let us take into
account the actual presence of limiting signal filters in all radar receivers. So,
the resulting signal in the working (usually low-frequency) region by the spectral
composition will be equivalent to the signal formed by the usual summation of
signals [8].
In the laboratory, as well as when working onboard with a low level of inter-
ference in the signal from the receiving antenna, it is possible to determine not
only the absolute value of the rate of the linear frequency change but also the
sign of that rate. This makes it possible to improve the quality characteristics
by performing a frequency shift only in the desired direction. This means that
harmonics corresponding to the second target will not be present in the output
signal of such simulator.
To achieve this result, the simulator is supplemented with a chirps sign detec-
tor, whose input is fed from the amplifier (Fig. 7). The signal from the output of
the chirps sign detector comes to an additional second input of the sign switcher
of the frequency shift value; for example, this may be the value “±1”, where
“+1” must correspond to “+∆f ” at the output of the sign switcher, and “–1”,
must match “–∆f ” at the output of the sign switcher.
Fig. 7. Simulator with the switcher of a sign of the frequency shift
When working on board of an aircraft, the level of interference and the level of
the useful signal can vary within wide limits. This can lead to an incorrect work of
29
the chirps sign detector and of the sign switch controlling the frequency shifting
device. Therefore, it is necessary to complete the measurement of the parameters
of the chirp signal by estimating the power characteristics of the signal from the
receiving antenna in order to adjust the parameters of the simulator. To provide
this, the simulator is supplemented with a Synchonizer-Detector (see Fig. 7).
The Synchonizer-Detector evaluates characteristics of the signal converted
into the mixer and sets the corresponding mode of operation of the simulator.
For example, it is provided the following way. If the signal level is insufficient,
the simulator mode is selected with periodic change of the modulation frequency
shift sign with equal intervals ∆t independently of the output signal of the linear
FM detector, and a gain value is given to the adjustable amplifier, for example,
in the range A = 2...3.
If the signal level is sufficient, the chirps sign detector must correctly monitor
the sign of the frequency change of the chirp signal. Therefore, such a mode of
operation of the simulator is established, in which the modulation frequency
shift is determined by the output signal of the chirps sign detector; here the gain
value is given to the adjustable amplifier, for example, in the range A = 1...1, 5.
The control of the gain value A allows one to improve the energy charac-
teristics depending on the mode of operation of the simulator by reducing the
level of the emitted signal of the target simulation by a factor of 2 with a correct
determination of the sign of the change in the frequency of the chirp of the input
signal. If there is no input signal with given/expected parameters or there is an
intermittent or pulse radar signal, the gain value A = 0 can be output to the
adjustable amplifier; this further improves the power characteristics.
To implement a delay line simulator device, modulators and adders can be
analog or digital. To improve the quality of simulation, the signal generation is
better to perform digitally on digital delay lines and modulators.
In [14], an example of the structure of a radar target simulator is given. The
structure provides formation of an equivalent sum of signals spaced along the
range of long-range target (or surface) elements. The unit uses the MC2301
PCI digital-signal memory-evaluation board based on the System-on-a-Chip
1879BM3 DSM developed by RC “Module” [13].
In accordance with [9] an additional amplifiers, attenuators, and possible
mixers can be chosen to implement the radar target simulator. For example,
suitable mixers, filters, and local oscillators should be used to downconvert the
high carrier frequency into the frequency range of the signal processing units.
To exclude the output signal of the transmitting antenna from entering to the
input of the receiving antenna, it is possible to use a circulator, strobe operation,
and / or spatial diversity of the antennas [1]. In stationary tests, it is possible to
connect cables directly to the radar system, that was investigated without use
of the antennas.
5 Conclusions
Peculiarity of the described solution for simulator constructing is that regardless
of the magnitude, direction, and combination of the signs of the rate of sawtooth
30
frequency change, two identical false radar targets are simulated. The first one
is the main target. It can be simulated at a range shorter than the range of the
radar carrier. The second target is assigned in range by 2∆Rmax , and with the
appropriate choice of the value of ∆f , this will not interfere with the correct
tracking for the main false radar target.
Acknowledgments. This work has been supported by the grant of the Ministry
of education and science of the Russian Federation (project No 8.2538.2017/4.6).
References
1. Perunov, Yu. M., Fomichev, K. I., Yudin, L. M.: Radioelectronic suppression of
information channels of weapons control systems (in Russian). 2nd edn. Radiotech-
nika, Moscow (2008)
2. Svy, K., Eastman, G., Lee, J.: Radio frequency signature augmentation system.
Patent application US 2008/018525 (2008)
3. Mills, K. G., Maxwell, T., Bergsagel, E. C., Richardson R. K.: Electromagnetic
target generator. Patent application US 5892479 (1999)
4. Method for deceiving a sonar or radar detector, and a decoy for implementing the
method. Patent application FR 2596164 (1987)
5. Mitchell R. L.: Electronic augmentation of radar techniques. Patent application
GB 2134740 (1984)
6. Jae-Hyun Choi, Jong-Hun Jang, Jin-Eep Roh: Design of an FMCW radar altimeter
for widerange and low measurement error. In: IEEE Transactions on Instrumenta-
tion and Measurement, vol. 64, issue 12, 3517-3525 (2015)
7. Bokov, A. S., Vazhenin, V. G., Dyadkov, N. A., Mukhin, V. V., Shcherbakov, D. E.,
Ponomarev, L. I.: Simulator of false radar target during linear frequency-modulated
signal probing (in Russian). Patent application RU 2486540 (2013)
8. Bokov, A. S., Vazhenin, V. G., Dyadkov, N. A., Mukhin, V. V., Shcherbakov, D.
E., Ponomarev, L. I.: Radar simulator aims with mainly long radiated signals (in
Russian). Patent application RU 2504799 (2014)
9. Sirotin, A. I., Mukhin, V. V., Nesterov, Yu. G., Dyadkov, N. A., Valov, S. V.:
Radar target simulator (in Russian). Patent application RU 2412449 (2011)
10. Vinitsky, A. S.: Essay on the fundamentals of radar in the continuous emission of
radio waves (in Russian). Sov. radio, Moscow (1961)
11. Salomasov, V. V., Shcherbakov, A. A.: Specific features of the simulation of a
reflected signal for chirp radar (in Russian). In: Izvestiya Vuzov, vol. 30, 8486.
Radioelectronics, Moscow (1987)
12. Beg, C.: A System Level FMCW RADAR Optimization for automotive powertrain
control application requirements. Waterloo, Ontario, Canada (2013)
13. Processor 1879BM3 (DSM) (in Russian). RC “Module”, Moscow. Available at:
http://module.ru/upload/files/vm3.pdf, last accessed 2017/10/10
14. Bokov, A. S., Vazhenin, V. G., Dyadkov, N. A., Mukhin, V. V., Shcherbakov,
D. E., Ponomarev, L. I.: Radar simulator aims with mainly long radi-
ated signals (in Russian). Patent application RU 2568899. Available at:
http://www.fips.ru/cdfi/fips.dll?ty=29&docid=2568899 (2015)