This paper is concentrated on designing a model of simulator, which can be used to simulate a radar signal reflected from a target or underlying surface. So it is useful to create and parameter tuning of the hardware-in-the-loop simulator. Here the key principle of signal channel mixing that aimed to simplify hardware architecture of the simulator is discussed. The dynamic variation of the simulating parameters of a radar scene (a surface type, an airborne altitude, etc.) also may be implemented.
1 i n E1 Ei En f1 fi … fn …
… … A(t) X(t)
Increasing the number of reflectors (the range rings should have the step value about or less than radar spatial resolution) significantly complicates the hardware for implementing the model with variable signal parameters. So to simplify the model it is advisable to group reflectors with similar parameters: the number of channels will be equal to the number of elements of the subdivision into areas of close frequencies or delays. In this case, the equivalence of the main characteristics and modeling dependencies and the formation of the reflected signal will be determined by the hardware and software capabilities of the chosen model implementation, as well as the actual resolving power of the radar meter with the corresponding parameters of the probing and received signals.
Immediate implementation on the extra high frequencies with the current development of technology is not possible, so the processing and generation of signals are performed at a low frequency in the working area of the DSP block frequencies [7, 8].
The use of signal mixing has shown the possibility to multiply the number of simulated bright points or significantly simplify the construction of the simulator. The possibility of coherent processing is preserved, because the phase of the signal being formed (imitating the reflection from each target point) is determined by the distance (delay) and the initial phase of the probe signal, and the possible amplitude fluctuation is averaged over several periods of modulation and scanning [7].
The basis of all high-speed digital systems of signal storage is DRFM (digital radio frequency memory) or digital
signal memory (DSM) [8, 12, 13]. One of the applications of such schemes is electronic warfare complexes. For
example, in [
The reflected signal is generated by irradiating a section of the underlying surface with the radio altimeter antenna, as shown in the figure 2. The elements of this section are at different distances from the radio altimeter, which leads to a difference in the delay of the reflected signals. So, the reflected (and formed) signals are random, since they represent the sum of the signals from the elementary areas of the reflecting surface section, each of which may have a random or controlled by the surface model parameters: amplitude, delay and phase.
The typical beat signal spectrum is shown in the figure 3. The main feature of the spectrum is the presence of discrete
components at frequencies that are multiple of the modulation frequency FM [
The envelope of the beat signal spectrum depends on the antenna pattern shape and the nature of the underlying
surface. The spectral component that has the minimum difference frequency, and will correspond to the true altitude of
the flight, and all the others “tail” spectral components will reduce the accuracy of the radar altimeter. The extent of this
inaccuracy is determined by the width of the beat signal spectrum. In practice the simplest method to evaluate the altitude
is measuring if the central of the spectrum, which is made in time domain by the counter of zero-crossing for the beat
signal [
G(f) 0 envelope (n-1)FM n FM
(n+1)FM
When the previous scheme was designed [10], we decided to develop an additional circuit capable of receiving and processing signals with different (variable) durations, rather than with the same ones, as was done in the previous case. In this case, it was necessary to modify the circuit of the cyclic counter with which the switch is smart controlled.
Some new elements were added: an additional counter and logical comparison device (each with an exit value of 1 or 0, corresponding to the truth, or a lie).
As a result, we developed a model that supports the processing of an incoming signal with none of multipliers in signal channels. With using variable durations, the form of the output signal with controlled power of different partial signals is obtained. It then will be proved by the spectrum of this signal in the frequency domain. The base part of our model of multipath signal propagation with the help of a simulator based on a cyclic switch with a variable duration in the Simulink environment is present in Figure 4.
The model of the simulator consists of different types of elements of the Simulink library, and main of them are: the sources of the chirp signal, the blocks with amplitudes (which are also responsible for the relative amplitudes of partial signals), and the cyclic switch, which is being controlled by a special law (smart rule). The cyclic switch that is suggested here is responsible for switching the signal channels with variable time durations.
As shown on the figures 4 and 5, relational operators are controlled by the cyclic switch shown on the fig. 5: the threshold (maximum value) of the main switch is 52 – it is a sum of the constants (2, 3, 5, 8, 13, 21, i.e. values from the blocks on the left of the fig. 4) and the cycle runs from 1 to 52. Each time step this value changes, therefore it is being compared to the following values from the adders in the relational operators. If 1-st value in each relational operator is more than current value of the switch, the relational operator outputs a value of 1 (truth), and this one passes to the main adder.
The possible values on the main adder’s output are: 1, 2, 3, 4, 5, or 6. So, this is the key information for the control port of the multiport switch. The switch connects one of its six ports and the signal passes through to the Dechirp mixer. The Dechirp mixer is a multiplier of two our complex signals.
So, the multichannel propagation model “transmit antenna reflecting object receiving antenna” can be represented as the model with one multitap delay line, and one switcher that is used instead of the combiner to simplify the hardware implementation. It is important to admit, that in our model, all the Doppler shifts are neglected to simplify the model. And, amplitude values simultaneously act with the delays in the corresponded signal channels. So, such signal processing does not include multiplication along all the signal channels; it can be described by the expressions: n j X (t) A(t i ) , i 1 if cnt t Ek , (1)
j1 k1 where i is the desired delay of the i-th partial signal;
A(t – i) is the delayed replica of the input signal; n is the number of partial signals (signal channels); Ei is the relative normalized amplitude that corresponds to the desired spectral power of the i-th signal; n cnt(t) is the value of digital counter that cyclically count from 1 to maximum value equal to Ek . k 1 if(bool) is the simple math function that gets 1 if the argument bool is true, or 0 otherwise. 5
The results of a model experiment are presented in Fig. 4-7. Here were used the following parameters: the spectrum width 1 kHz; the simulation stop time is 80 ms; 6 signal channels (suitable for simplified modeling).
The processing of the cyclic switch is presented in the figure 6. It is important to note that all the six steps are different in durations, and the durations are proportional to the relative normalized amplitude values defined by Ei. In the figure 7 the input signal (in-phase component) is presented: After that, we can obtain a form of the beat signal, an example of which is shown in figure 8: Next, we get a spectrum view that is presented similar to a fence in figure 9: And finally, the zoomed single spectrum part is presented in figure 10. One can see that power values of the spectrum correspond to the values defined by Ei in the model experiment.
The characteristics of the signals simulated in the Simulink in the time and spectral domain are close to theoretical and expected features [9, 11]. Therefore, the results will correspond to the desired experimental signals for simplified tests and analyzing of altimeter’s operation while it works with multipath propagation signals.
The number of signal channels can be increased by a number of model improvements. What is more, a facet model [8], which allows us to form reflected signals from typical terrains, may be implemented in our model. The model allows us to evaluate software control signals for the simulator which may be constructed with same signal processing ideas. 6
In this paper, we have presented the simulator for radio altimeters, especially for extended radar objects. At the beginning we designed the method for building the simulator for multipath propagation of radar signals with variable durations of signal’s switching intervals. Then the simulator scheme was formed. It gives us an opportunity to form reflected signal by switching the hardware channels with different signal parameters that can be changeable in real time [8, 11]. After this, we have presented some experimental results, which in common correspond to practical and theoretical results [8]. Formed characteristics of the simulated signal in the time and spectral domain are close to the theoretical and real flight results in similar conditions for simple natural surfaces.
The relevance of this project is rather significant, because we finally managed to simplify the hardware structure of the simulator via presenting this device in the MATLAB. Moreover, it is important that the solution mentioned above is working properly with mixing intervals, which are unequal in duration [10] of a simulator implementation.
The obtained results can be helpful in the development and improvement of hardware-in-the-loop simulators for checking and verification of various autonomous airborne radars or altimeters with frequency modulation, as well as for other radar systems with continuous or long emitted signals.
Acknowledgments This work has been bankrolled by the grant of the Ministry of Education and Science of the Russian Federation (project № 8.2538.2017/4.6). 4. A. S. Bokov, V. G. Vazhenin, N. A. Dyadkov, A. A. Iofin, V. V. Mukhin. Possibilities of studying the accuracy characteristics of airborne radio-altimeter systems based on the simulated reflection signals (in Russian). Reliability and quality of complex systems, 1(13) :86{93, 2016. 5. S. Z. Zubkovich. Statistical characteristics of radio signals reflected from the Earth's surface. Sov.radio, Moscow, 1968. 6. T. A. Lepekhina, V. I. Nikolaev. Experimental determination of spaceborne SAR radiometric resolution. The 22nd
International Conference “Microwave & Telecommunication Technology”, Sevastopol, 1009{1011, 2012. 7. A. S. Bokov, V. G. Vazhenin, N. A. Dyadkov, A. A. Iofin, V. V. Mukhin. Check and study the accuracy characteristics of the airborne radio altimeters. Reliability and quality of complex systems, 3(16) :78{82, 2016. 8. V. G. Vazhenin [and others]. Seminatural modeling of onboard radar systems operating on the Earth's surface: textbook (in Russian). UrFU, Ekaterinburg, 2015. 9. A. S. Bokov, V. G. Vazhenin, N. A. Dyadkov, V. V. Mukhin, D. E. Shcherbakov, L. I. Ponomarev. Radar simulator aims with mainly long radiated signals (in Russian). Patent application RU 2568899, 2015. 10. A. S. Bokov, S. A. Margilevsky, V. G. Vazhenin. Simulation of extended radar objects on the basis of several signal channel mixing. USBEREIT, IEEE, 249{252, 2018. 11. A. S. Bokov [and others]. Experimental study of seminatural simulation of a radar channel (in Russian). Reliability and quality of complex systems, 3(11) :91{98, 2015. 12. J. Yang, Y. Li, Y. Zhang. Realization of novel DRFM jamming source based on AFB-SFB. 5th International
Congress on Image and Signal Processing, CISP, 2012. 13. Z. Peng. Realization of DRFM radar target simulator based on general instruments. IET International Radar Conference, 1841{1846 China, 2015.