RU2230337C2 - Signal processing device built into radar with phase arrays - Google Patents

Abstract

FIELD: radiolocation. SUBSTANCE: invention can find use in measurement of angle of elevation coordinates of radiation sources in radar carrying phase array under action of intensive active jamming and other interference, in improved resolution of signals displaying various intensity by angle of elevation coordinate. Signal processing device built into radar with phase arrays includes N antenna elements, N reception channels each incorporating SHF amplifier, mixer, matched filter, quadrature phase detector, electron key and storage, local heterodyne, coherent heterodyne, commutator. Device is supplemented with quadrature analog-to-digital converter, first quadrature spectrum analyzer, meter of signal parameters, threshold unit, permanent storage, quadrature delay unit, first quadrature multiplier, quadrature rejection filter, second quadrature multiplier, second quadrature spectrum analyzer and delay unit. EFFECT: enhanced noise immunity of signal processing device of radar with phase array from active jamming. 1 cl, 10 dwg

Description

The invention relates to radar and can be used to measure the elevation coordinate of radiation sources in radars containing a phased antenna array (PAR) under conditions of intense active noise and other noise. It is possible to use the device to improve the resolution of signals having different intensities in elevation coordinate.

It is known that the processing of signals in radars with headlamps can effectively solve a number of complex radar problems [1]. Widespread use of the device based on the phased array is found in those cases when it is necessary to quickly measure the elevation coordinate of airborne objects (VO). In this regard, as a prototype of the selected device signal processing in phased antenna arrays "[2].

This device contains antenna elements, channel receivers, electronic keys, phase detectors, storage devices, a switch and a spectrum analyzer. The essence of the device consists in the formation of a special signal corresponding to the distribution of signals in the vertical section of the PAR aperture and its spectral analysis. For this, in each n-th channel of the PAR (n∈ [1, N]), after amplification and optimal processing in the channel receivers, the signals are sent to electronic keys, in which signal samples of duration ≤T _and / N are simultaneously taken. Further, these samples are stored and with the help of an electronic switch a single signal is created from them. In this signal, the samples are arranged regularly and in such a way that samples with numbers n-1 and n + 1 are located near the sample with number n. As a result of this, a signal of duration ≤T _{and is} formed after the switch.

It is known that the phase difference of the signals in the adjacent elements of the PAR is

Here Δd is the distance between adjacent PAR elements in the vertical plane;

λ is the radar wavelength,

ε is the elevation angle of the radiation source.

Since the sequence of samples in the signal at the output of the switch and the location of the signals in the headlamp are the same, and the changes in the initial phases of the samples are determined (1), it can be shown that the carrier frequency of such a signal depends on the angle ε

Spectral analysis allows us to distinguish between signals with different values of ωε, and hence with different angles ε (2). Measurement of ωε using a spectrum analyzer allows you to determine the elevation angle (2) of radiation sources located in the VO.

The use of the device [2] does not allow to achieve the goal under the conditions of active noise interference. The fact is that the use of auto-compensators [4] allows you to protect the device from active noise interference (ACP), only in the area of the side lobes of the directivity pattern (NAM) of the PAR in the azimuthal plane. If the ACP source is received by the main lobe of the HEADLAM, then the correlation of the interference in the main and compensation channels of the autocompensator is reduced and the interference is not compensated. The use of polarization selection for compensation improves the security of the RDS from interference. However, it is effective only in certain special conditions, when one source of noise interference is used to radiate them with different polarizations. Obviously, these conditions are easily eliminated by using independent sources of ACP or by rotating the plane of interference polarization.

Thus, the protection of the considered signal processing device from ACP in the main lobe of the azimuthal daylight phased array is practically not provided.

Given that the ratio of the ACP intensities and the usefulness of the signal can reach 60-70 dB, it is clear that under these conditions it is impossible to measure the angular coordinate of those HEs whose azimuthal coordinate practically coincides with that of the ACP director. This feature of the known device is a serious disadvantage of it.

The aim of the present invention is to improve the security of the signal processing device in the radar with the PAR from active noise interference received on the main lobe of the azimuthal beam. It can be argued that the claimed device will still allow signals with very different (up to ≈60 dB) intensities. Therefore, in the inventive device will improve and resolution in the angular coordinate ε when receiving signals of various intensities. This significantly increases the efficiency of radar with headlamps and can ensure its operability under the influence of other types of interference, such as passive and pulsed.

This goal is achieved by the fact that in the known device, taken as a prototype and containing antenna elements, channel receivers, phase detectors, electronic keys, memory devices and a switch with a spectrum analyzer, according to the invention, a signal parameter meter connected in series with a threshold is connected in series with said spectrum analyzer a device with read-only memory, the outputs of which are connected to the multipliers, the first directly and the second through a delay device, inter for which the notch filter of low frequencies is included. The input of the first multiplier is connected to the output of the delay device, the input of which is connected to the input of the spectrum analyzer and the output of the analog-to-digital converter, the input of which is connected to the switch; the output of the second multiplier is connected to a newly introduced spectrum analyzer, the output of which is connected to the signal parameter measuring device mentioned above. Spectrum analyzers, an analog-to-digital converter, devices for measuring signal parameters, delays, as well as read-only memory are connected to a source of clock pulses to ensure the necessary sequence of their operation.

As the analysis showed, the features that distinguish the claimed device from the prototype are its permanent storage device, which is controlled through a threshold device, signal multipliers with a notch filter and a second spectrum analyzer with a signal delay device. Thus, we can conclude that it meets the criterion of "novelty."

The use of notch filters is known and used, for example, in SDC devices [7]. Usually they are performed in the form of linear filters with a finite pulse transition function (IPF) or an infinite IPF in linear feedback systems. In the inventive device, the use of such filters is possible. However, the presence of transients determines the gradual transition to the maximum interference suppression mode. In this case, the time constant should be chosen from compromise considerations that provide, on the one hand, greater suppression and, on the other hand, preserve the signal duration. In the claimed device can be used interpolation notch filter based on the spectral representation of the notched signals [8].

Analysis of the spectrum of the input signal in a certain limited area allows reproducing with the necessary accuracy the original signal (interference) and eliminating it from the input signal. At the same time, compromise requirements for the parameters are removed and the suppression of interference is maximized.

The implementation of notch filters is optimal for low-frequency signals [8]. Therefore, in the inventive device, before the notch, the frequency of the ACP signal is converted to the zero frequency region where the notch is performed, and the frequency of the signals remaining after the notch is restored.

The second feature of the claimed device is the use of read-only memory (ROM) controlled by a threshold device. The ROM stores the information (coefficients) originally entered into it of all possible values of the measured parameter, i.e. angle ε. The selection of the corresponding coefficients from the ROM by the results of measuring the ε coordinate of the ACW source ensures the conversion of the frequency of the delayed ACW signals to the zero region. Moreover, the threshold device "allows" this conversion only if the intensity of the input signals, i.e. essentially ACP signals are large enough. Thus, the device responds to the appearance of only a hazardous ACP in intensity and suppresses (compensates) it. In the case of small ACW signals, they are not compensated, the device operates in the usual way, and the processing of VO signals is provided due to the selective properties of the device.

Another feature of the claimed device is associated with the presence of a second spectrum analyzer and a delay device for the signals arriving at it. This feature is determined by the finite duration of the signals (≈Т _and ) and the time spent on spectral analysis in the spectrum analyzer. In this regard, information about the appearance of the ACP in the input signal is delayed relative to the beginning of the signal. Under these conditions, a signal delay and a second spectrum analyzer are necessary for ACP rejection and repeated spectral analysis. The latter is necessary so that the analysis of input signals is not interrupted. In the inventive device, the total signal delay in the channel of the second analyzer is negligible and can reach (2-3) T _and .

The considered features of the claimed device are significant, because only if available can the objective of the invention be achieved. Indeed, signal frequency conversion provides the possibility of efficient ACN rejection while maintaining the signal duration, and restoration of the signal frequency in front of the second spectrum analyzer allows you to save a single frequency (elevation) scale of both spectrum analyzers. The use of a threshold device makes it possible to fix (detect) ACP signals of “dangerous”, high intensity and include a channel for rejecting these interference and secondary spectral analysis to detect and measure the coordinates of HE “covered” by the ACP director.

Thus, in the claimed device, the purpose of the invention can be achieved only on the basis of the totality of all the features inherent in it and absent, as signs inherent in it and absent, as shown by a comparative analysis, of known devices.

Only the combined use of the features of the prototype and the features indicated in the characterizing part of the claims ensures the appearance of new properties of the claimed device and thereby confirms the compliance of the claimed device with the criteria of "novelty" and "significant difference".

The stated essence is illustrated by graphic material. Figure 1 shows a block diagram of a device adopted as a prototype, and figure 2, 3, 7 of the claimed. 1-3, 7 the following notation is used:

1 - antenna element,

2 - channel microwave amplifier

3 - mixer

4 - local oscillator

5 - matched filter,

6 - two-quadrature phase detector

7 - coherent local oscillator,

8 - electronic key,

9 - storage devices

10 - switch

11 - analog-to-digital Converter,

12 - spectrum analyzer,

13 - meter signal parameters,

14 - device delay signals

15 is a threshold device

16 - read-only memory device

17 - multiplier,

18 - notch filter,

19, 20 - input signal synchronization,

21 - output device

22 is a subtractive device,

23 - integrator

24 - delay line

25 - adder

26 - balanced mixer,

27 - balanced modulator,

28 - controlled local oscillator.

Figure 4 shows the structure of the PAR, explaining the properties of the received signals. In figure 4, the following notation is used:

1, 2, .. n, .. N - numbers of PAR elements in the vertical plane,

O is the surface of the earth

ε is the elevation angle of the source of the received signal,

d ₁ - the distance from the surface O to the first element of the PAR,

Δd is the distance between adjacent elements of the PAR,

Δr is the difference in the signal path for neighboring PAR elements.

Figure 5, 6, 8 shows a plot of signals explaining the operation of the claimed device. The following notation was used:

T _sq. - the quantization period (taken samples) of the signals in the PAR elements,

U _s - envelope of the signal reflected from the VO in one of the channels of the PAR,

U _ACP - the envelope of the ACP signal in one of the headlamp channels,

T _{sq. 1} - the repetition period of samples from the signals received by the PAR elements in the formed single (combined) signal,

φ _s , φ _ACP - phase structure of the combined signal, respectively, for the cases of the VO signal and the ACP signal,

20 - clock pulses,

S _i - signal in the i-th element of the range,

T _about - the delay of the signals during processing,

T _3s - the total delay of the processing result,

10-18 - the number of elements of the claimed device,

,

- соответственно модули сигналов считываемых коммутатором 10 и их спектров,

,

- respectively, the signal modules read by the switch 10 and their spectra,

ω _No , ω _si are respectively the carrier frequencies of the ACP signal and the BO signal,

ΔB ₁ - the relative change in the levels of spectral density in the main and first side lobes,

ΔВ ₂ - octave relative change in spectral density levels in the side lobes,

ω _max - the maximum frequency of the signal, determined by the discrete structure of the PAR,

K _f - frequency response of the notch filter.

The claimed device consists of N antenna elements 1. After each element 1, a channel microwave amplifier 2, a mixer 3 with a connected local oscillator 4, a matched filter 5, a two-quadrature phase detector 6 with a connected coherent local oscillator 7, an electronic key 8 and a storage device 9 are connected in series. Storage devices 9 of all N channels of the HEADLIGHT are connected to a switch 10, to the output of which an analog-to-digital converter 11 is connected. To its output 11 is connected a spectrum analyzer 12 connected to a param meter trov signal 13, the output of which 21 is the output of the entire device. In parallel with the spectrum analyzer 12, a signal delay device 14 is connected, the output of which is connected to a series-connected multiplier 17, a notch filter 18, a second multiplier 17 and a second spectrum analyzer 12, the output of which is connected to a signal parameter meter 13. The outputs of the signal parameter meter 13 are connected: one - with a threshold device 15, the other with read-only memory 16, the two outputs of which are connected respectively to the second inputs of the multipliers 17, and the first multiplier 17 is directly but, for the second, through the second signal delay device 14. The clock pulses 19 are connected to the switch 10, the ADC 11, the signal delay devices 14, the read-only memory 16 and the notch filter 18. The synchronization pulses 20 are connected to both spectrum analyzers 12, the signal parameter meter 13 and the notch filter 18. The phase detector 6 and all subsequent devices, except for the signal parameter meter 13 and the threshold device 15, are two-quadrature.

The device operates as follows. The signals received by the I PHA elements in each of the N channels are amplified by microwave at 2, converted by the local local oscillator 4 and mixer 3 to an intermediate frequency, where they are filtered using a matched filter 5. Next, the signals are detected in a two-quadrature phase detector 6, to which the coherent oscillator oscillates 7. As a result, two video signals are formed, which are projections of the signal vector onto the components of the oscillations of the coherent local oscillator shifted by π / 2 7. Using electronic keys 8 controlled by a clock E 20, of video signals simultaneously in all the channels is obtained N FAS short sampling. These samples are stored in devices 9 and read out of them sequentially by the switch 10. At the same time, simultaneous samples from the signals of N elements of the PARS are converted into sequential ones with a regular sequence. In this case, the first is sampling from the extreme element of the PAR and then all subsequent ones.

. При этом всего в устройстве 16 необходимо иметь 2 N² (5-10)² значений фазы, образующих комплексные коэффициенты

. После преобразования в умножителе 17 частоты сигналов АШП к нулевой частоте (с использованием комплексно-сопряженных коэффициентов ПЗУ) происходит режекция их в режекторном фильтре 18. Для этого в режекторном фильтре анализируется спектр сигналов в небольшой области около нулевой частоты и по этому спектру воссоздается сигнал нулевой частоты, содержащийся во входном сигнале. Далее он вычитается из входного сигнала, а оставшийся после компенсации сигнал поступает на выход режекторного фильтра 18 и далее на вход второго умножителя 17. В этом умножителе используются Фурье-коэффициенты, комплексно-сопряженные тем, что использовались в первом умножителе 17. Поэтому после второго умножения частота сигналов восстанавливается к исходному ее значению, существовавшему до первого умножения. В режекторном фильтре 18 происходит задержка сигналов в процессе режекции сигналов АШП. Поэтому коэффициенты ПЗУ 16 для второго умножителя 17 задерживаются в устройстве задержки 14, включенном между устройствами 16 и вторым умножителем 17. После второго умножителя 17 сигналы поступают на второй анализатор спектра 12, выполненный аналогично первому. В этом анализаторе производится спектральный анализ сигналов, оставшихся после компенсации АШП и далее в 13 происходит определение частоты (угла места) их (2). Информация об этом из измерителя параметров сигнала 13 поступает на выход 21 всего устройства.The clock cycle 20 is selected in accordance with the Kotelnikov theorem and is equal to T _sq ≈ 1 / Δf _n ≈ T _and is the signal duration. The duration of the signal samples is less than or equal to T _n / N. Therefore, after the switch 10, signals are generated whose duration does not exceed T _and . These signals are fed to the ADC 11 and each sample is converted to digital form based on a binary digital code. Next, the signals are fed to a spectrum analyzer 12, where they are analyzed by harmonic components, i.e. Fourier analysis. An analysis of 12 produces a signal in the form of a spectrum modulus. Therefore, two quadratures of the input signal in the analyzer 12 are converted to a single spectral (spectrum module) signal. This signal enters the signal parameter meter 13, where the frequency (carrier) and its amplitude are measured. The signal frequency value is associated (1, 2) with the elevation angle ε of the signal source and its measurement is the most important task of the claimed device. Information about the measured ε is output to the entire device 21. In addition, information about ε is fed to the input of the read-only memory 16. The result of measuring the signal amplitude is transmitted to the threshold device 15, where it is compared with a predetermined threshold value. The purpose of the comparison is to determine the “dangerous” (threshold) level of the ACP signal at which it is impossible to detect useful signals from this azimuthal direction — the direction to the ACP source. For example, if the threshold signal at the output 12 is ρ _p and when the noise level at the output is 12≥ρ _p / 2, the measurement and detection becomes unreliable, then, taking into account the level of the side lobes in the vertical section of the PAR of the claimed device, β _l , the threshold level for 15 should be β _p ≈ ρ _p / 2β _L. If the signal at the output 12 (13) containing the ACW exceeds this level, then a signal is received from the threshold device 15 to the ROM 16, allowing its signals to be included in the multiplier 17 and the delay device 14. In the ROM 16, the coefficient values necessary for converting the frequency of the input signal (2) to zero frequency. Information about the signal frequency obtained in 13 is used to select the necessary coefficients from 16. The selection rule is that out of 16, the frequency (coefficients) closest to the measurement result in 13 is selected. The concept of frequency, as applied to the case under consideration, can be clarified in the sense that one should keep in mind a set of discrete values of complex Fourier coefficients. Moreover, the discrete values are samples from a continuous signal with a carrier of type (2). It is known [1] that the minimum discrete value on the elevation scale is Δε _min ≈ λ / 2L _A (L _A = N · Δd is the size of the PAR aperture). However, when using such a discrete in ROM 16, the conversion to zero frequency will occur rather roughly and this will necessitate the use of a very complex filter 18 — a high-order filter. Therefore, it is advisable to use 16 smaller discretes

. Moreover, in device 16, it is necessary to have 2 N ² (5-10) ² phase values forming complex coefficients

. After converting the frequency of the ACP signals to the zero frequency in the multiplier 17 (using the complex conjugate ROM coefficients), they are notched in the notch filter 18. For this, the signal spectrum is analyzed in the notch filter in a small area near the zero frequency and the zero-frequency signal is recreated from this spectrum contained in the input signal. Then it is subtracted from the input signal, and the signal remaining after compensation is fed to the output of the notch filter 18 and then to the input of the second multiplier 17. This multiplier uses Fourier coefficients complex conjugate to those used in the first multiplier 17. Therefore, after the second multiplication the frequency of the signals is restored to its original value, which existed before the first multiplication. In the notch filter 18 there is a delay of signals during the rejection of ACP signals. Therefore, the ROM coefficients 16 for the second multiplier 17 are delayed in the delay device 14 connected between the devices 16 and the second multiplier 17. After the second multiplier 17, the signals are fed to the second spectrum analyzer 12, which is similar to the first. In this analyzer, a spectral analysis of the signals remaining after the ACW compensation is performed, and then at 13, their frequency (elevation angle) is determined (2). Information about this from the signal parameter meter 13 is output 21 of the entire device.

Elements of the claimed device operate synchronously. The main clock frequency is the reading frequency of the samples of the signals of the HEADLIGHTER 19. In this case, during the duration of the signal T _, N samples must be read and, consequently, the repetition period of samples T _sq. 1 = T _and / N. At the same speed, the coefficients are sampled from the ROM 16 and the operation of the signal delay devices 14 and ADC 11 is controlled. To control the electronic keys 8, select samples from the PAR elements by spectrum analyzers 12 and signal parameter meter 13 using more rare (N times) clock pulses 20.

Figure 5 shows a plot of stresses explaining the operation of the claimed device. The plot of U _s shows the signal module in one of the headlamp channels. As an example, a simple harmonic signal with a rectangular envelope is selected. After the matched filter 5, this signal has an envelope of a triangular shape. As can be seen from the plot, two samples will be taken from the signal. Corresponding quantization pulses (U _quarter ≡ 20) are marked. The ACP signal module is depicted in the following diagram. In contrast to the signal, the ACW is continuous, and its amplitude is random. Therefore, ACP samples are formed over a longer period of time (range).

The following diagrams, performed on a larger time scale, show the signals after switch 10. They form a regular sequence of samples of N elements of the PAR in the increasing, for example, the order of their numbers. The sampling period T _{sq. 1} during the formation of a single signal is 10 less and a multiple of the sampling period T _sq. . On the 5th and 6th diagrams, respectively, are shown the sequences of amplitudes of the signal samples and their initial phases. In accordance with (1), the values of the initial phases are determined by the path difference between the PAR elements, and their change occurs monotonously. The last two plots show similar ACP signals. We see that for the ACP, the change in the amplitudes of the samples is random, and the sequence of initial phases has a random phase of the first sample. The sequence of the initial phases is regular because it is determined by the angle ε of the ACP source. Figure 5 shows the signals for the case when there are no reflections from the surface of the earth. The existence of such a signal (figure 4) leads mainly to modulation of the amplitude of the signal U _s formed by alternating samples of PAR elements [1] and is not essential for the operation of the inventive device. The diagrams in Fig.6 give an idea of the sequence of the signal in the elements of the claimed device. We see that the signal in the ith range element, i.e. The i-th samples from the PAR elements, (S _i ) occupies almost the entire quantization interval. This signal is generated by devices 10, 11. In the spectrum analyzer 12, the signal, as is known, cannot appear before the input signal S _i ends. Therefore, the result of the analysis is observed in the next quantum. The parameter meter and threshold device (13, 15) process the signal in the same quantum of distance. From 16, the selection of coefficients is carried out only after the end of the measurement process in 13, i.e. in the next quantum (the second from the start of the signal S _i ). In the same quantum, the signal in the multiplier 17 is multiplied. Therefore, the signal for this multiplier is delayed at 14 by two quantization intervals (T _o ). In filter 18, the signal is delayed by one quantization interval, because filtering is associated with the determination of the special spectrum S _i . This signal is processed in the second multiplier 17 (2). Fourier coefficients are supplied to its second input from 16. Since they are generated earlier, their delay in the second device 14 by T _{about is} necessary. As can be seen, the value of T _about equal to two quantization intervals. Next, the signal is analyzed in a second spectrum analyzer and this also takes time in one quantization interval. We see that the total time _spent is T _3s , equal to four quantization intervals. These costs are negligible, constant and can easily be taken into account when processing signals in the radar.

The claimed device is an implementation option based on digital technology. This is due to the capabilities of its elemental base, high accuracy of operations, their wide range and high technology equipment. However, the claimed device, like the prototype, can be performed on the basis of analogue technology. 7 shows a functional diagram of such a device. We see that it is identical to the device depicted in figure 2. When switching to an analog device, delays are replaced by delay lines 9 → 24, 14 → 24. The ADC is excluded 11. The ROM is replaced by a controlled (in frequency) local oscillator 16 → 28. The multipliers are respectively replaced by a balanced mixer and modulator 26 and 27. Signal processing in an analog device does not require two quadrature. Only the notch filter 18 remains two-quadrature. All these features are reflected in the diagram of Fig. 7. The operation of this device is similar to digital and does not require special explanations.

Let us consider briefly the main signal ratios in the elements of the claimed device, which determine the efficiency realized by it. For simplicity, we will use the analog waveform. Let the ACP signal and the useful signal be received at t∈ [0-T _and ]

,

- соответственно векторы АШП и полезного сигнала;here

,

- respectively, the ACP and useful signal vectors;

ω _N , ω _s are respectively the frequencies of these signals, related by relation (2) with elevation angles;

T _about - the duration of the signal (quantization interval)

The spectrum analyzer determines:

оказываются обычно больше полезного сигнала и "маскируют" его. В этих условиях влиянием сигнала на помеху можно пренебречь и измерить частоту помехи обычным для прототипа способом - по положению максимума спектра. Далее сигнал (3) преобразуется с гетеродинным напряжением, частота которого определяется частотой АШП, т.е. ω_r=ω_N. При этом получается сигнал следующего вида:Since this operation is linear, and the interference is much larger than the signal, the latter is not detected in the spectrum. This is due to the side lobes of the spectral signals [10], which, when

usually there are more useful signal and “mask” it. Under these conditions, the influence of the signal on the interference can be neglected and the interference frequency measured in the usual way for the prototype - by the position of the maximum spectrum. Next, the signal (3) is converted with a heterodyne voltage, the frequency of which is determined by the frequency of the ACP, i.e. ω _r = ω _N. This produces a signal of the following form:

Next, it is necessary to edit the ACP signal of zero frequency and at the same time save the useful VO signal having a different frequency (3.1). To do this, you can apply a filter based on the interperiodic compensator of the SDS system (10, p. 48). However, such notch devices, being linear devices, have a pulse transition function with a finite duration. Therefore, during rejection, transients occur that lead to a compromise between the suppression of ACP and a decrease in sensitivity with respect to useful signals.

However, in the inventive device, a nonlinear filter based on the spectral representation of the signals can be used for rejection. At the same time, there is a constant signal delay in the filter, and transients similar to those observed in linear filters are absent. In this regard, it is possible to deeply reject the ACP signals and the almost complete absence of loss of sensitivity in relation to useful signals. Since the use of nonlinear filters is associated with more complex processes, calculations are presented below that explain them and give estimates of achievable results.

, имеющего отличающуюся от нулевой несущую частоту, в значения спектральных коэффициентов незначителен. Следовательно эти коэффициенты, как и воссоздаваемый сигнал U_s2 определяются в основном сигналом АШП. Так, если используется лишь полином нулевого порядка, т.е. P_o(t)=I, то спектр:The notch filter decomposes the signal U _s1 into a spectrum according to power orthogonal polynomials [8] to the second or fourth degree. The limitation of the number of polynomials is associated with the need for more accurate reproduction of ACW signals in the conditions of small angles ε of their reception, when the effect of the signal reflected by the ground affects [1]. For the remaining angles ε, we can restrict ourselves to polynomials of degree zero and degree one. Further, the signal U _{s2 is} reconstructed from the polynomial spectrum using the inverse spectral transformation. Since the polynomial spectrum was determined by the initial polynomials (not higher than the 4th degree), the signal contribution

having a different carrier frequency from zero, the values of spectral coefficients are negligible. Therefore, these coefficients, as well as the reconstructed signal U _{s2, are} determined mainly by the ACP signal. So, if only a polynomial of order zero is used, i.e. P _o (t) = I, then the spectrum:

When a signal is formed based on spectrum (4), we obtain:

. Из соотношения (4), в котором определен сигнал

, следует, что

. После преобразования частоты этого сигнала в умножителе, включенном после режекторного фильтра, восстанавливаем прежние частоты сигналов:After subtracting this signal from the converted input (U _s1 ), we have

. From relation (4), in which the signal is determined

, follows that

. After converting the frequency of this signal in the multiplier included after the notch filter, we restore the previous signal frequencies:

легко обнаруживается и измеряется его частота.As a result, the signal U _s4 contains the useful signal and the "remainder" from the ACP, the value of which does not exceed the useful signal. In the spectral analysis of this signal in the second spectrum analyzer, the signal

its frequency is easily detected and measured.

Thus, it is possible to detect and measure the frequency (satellites) of signals in a radar with a phased array in conditions when a useful signal is received in the main lobe of the radiation pattern simultaneously with the ACP. Consideration of the principle of operation of the claimed device shows that it performs rejection of intense interference based on the use of spectral differences of the useful signal and the ACP. Indeed, the interference signals are converted into the region of frequencies being cut in the filter, and the BO signals having a different elevation coordinate, and hence the frequency, are outside the notch zone, stored at the output of the filter 18, and observed during further processing. The spectral composition of all signals processed in the device and significant from the point of view of their selection is determined by the structure of samples from the PAR [2]. In this regard, the ACP samples for each distance element are correlated, since the effective vertical size of the PAR aperture (N · Δd · sin ε) usually does not exceed the radar resolution element in range. In view of the foregoing, it should be borne in mind the possibility of a similar selection in the claimed device and intense interference of a different kind. For example, the passive noise received by the considered PAR is correlated and has a frequency color, determined by their elevation coordinate. Therefore, their recession in the claimed device will allow detecting HE in this distance element at a different angle ε.

The implementation of the claimed device is possible on the basis of well-known technical solutions using the existing element base. Its elements such as ADCs, delay and multiplication devices, and PCRs are basic and basic in digital technology and are used directly in the form of units functionally designed and defined by standards [11, 12]. The spectrum analyzer, the threshold device and the signal parameter meter are devices whose theory and implementation methods have been studied in detail. Spectrum analyzers are performed using multipliers, adders and storage devices containing the coefficients of the spectral basis functions. The meter parameters of the signal 13 is necessary to determine the amplitude and frequency of the signal, which corresponds to the ε - coordinate IN or the source of interference. As shown, for example, in [4], the measurement of the angular coordinate reduces to determining the position of the signal maximum at the output of the spectrum analyzer. The structure of such meters is known and studied in detail [4, p.268].

As a notch filter, it is possible to use notch filters similar to those used in SDC systems (10, p. 482-487, 497-501). It is clear that the signal delay time (recurrence period) is determined by the periodicity of the simultaneous sampling taken in the PAR elements (T _sq1 , Fig. 5).

As indicated earlier, it is more efficient to use notch based on interpolation processes using spectral signal transformations. In connection with some specifics, we explain the operation of the notch interpolation filter. A diagram of it is shown in Fig.3.

The filter is two-quadrature and in each quadrature contains 2-4 multipliers 17, the signal inputs of which are connected and form the quadrature inputs of the filter 18. The outputs of the multipliers 17 are connected to integrators 23, after which the electronic keys 8, memory devices 9 and the second multipliers 17 are connected in series. Second the inputs of the multipliers 17 are interconnected and with the outputs of the ROM 16. The outputs of the second multipliers 17 are combined by a summing device 10, the output of which is connected to a subtractor 22. The second input 22 through the delay device and 14 is connected to first inputs of the multipliers 17. The output of the subtractor is the output of the notch filter 18 in each quadrature. The clock pulses 19 and 20 are connected respectively to ROM 16, delay devices 14, electronic keys 8 and integrators 23.

The notch filter works as follows: video signals containing an ACW at zero frequency and a useful signal of a certain frequency are fed to the multipliers 17. Reference oscillations are supplied to the second inputs of the multipliers 17 from the read-only memory 16, the shape of which is determined by power orthogonal polynomials

(Legendre, Chebyshev, etc.) [8]. After multiplication, the signals are integrated into 23 and the integration result using the electronic key 8 at the time corresponding to the leading edge of the pulse 20 is transferred to the storage device 9. Integration at 23 and storage at 9 is performed on the quantization interval T _sq. . Preliminarily, the integrator 23 is reset to zero at the time of the trailing edge of the clock 20. In the filter, low-frequency (not higher than 4th order) polynomials are used. Spectral coefficients of 9 are stored per T _square. and multiplied by the same reference oscillations in the second multiplier 17 and after the adder 10, the "input" signal is recreated. If we take into account that the reference oscillations are low-frequency and not more than four are used, then only signals with a carrier frequency of about zero participate practically in the formation of the signal after the adder 10. This is a consequence of the main property of spectral transformations [10]. Indeed, for the exact reproduction of signals in a spectral pattern, it is necessary to decompose them into an infinite spectrum. In our case, this would require the use of an infinite set of support functions - from zero to infinite degree. If the spectrum is limited in frequency, this makes it impossible to represent rapidly changing signals and preserves only the low-frequency components of the signal. In our case, such a signal is the converted ACP signal. The signal received from the VO with a different angle ε has a different carrier frequency [2], different from zero. Therefore, after the adder 10, only the ACW signal is recreated. Then it is fed to a subtractor 22, to the second input of which, through the delay device 14, an input signal is supplied containing both the ACW and the BO signal. After subtracting at 22, the ACP signal is compensated and only the BO signal is received at the filter output. The required signal delay of 14 is determined by the delay of the signal after 10, which, as can be seen from the above, is determined by the integration time, i.e. the value of T _and (the duration of the quantization interval). The reference functions are known and pre-recorded in the ROM 16. The clock pulses 19 and 20 provide their reading with the necessary speed and frequency.

The implementation of the notch filter 18 is possible using the main known and existing devices from the nomenclature of the digital element base [11, 12]. Indeed, adders, memory devices, comparators, shift registers, and read-only memory devices should be used to create a notch filter. All of them are defined by uniform standards and are given in reference books on general-purpose microcircuits [11 p. 489 - adders, p. 490 - shift registers, p. 494 - memory, p. 492 - multiplexers, etc. 12] with a detailed indication of the modes feeding and control voltages and parameters of consumer loads. The implementation of the claimed device based on them does not require inventive thought.

Thus, it should be considered that the goal of the present invention in the claimed device is achieved. As a result of this, in the radar with the PAR it is possible to eliminate the ACP received by the main lobe of the azimuthal radiation pattern of the PAR. In this case, HE signals become detectable, the elevation coordinate of which differs from that of the ACP source.

We can assume that the claimed device implements the possibility of resolving the signals at elevated coordinates, the intensity of which varies by several orders of magnitude. Distinguishing such signals by conventional means in the spectral and time domains is practically impossible.

It should be noted that in the claimed device, rejection (interference compensation) is based on an aperture analysis of signals. Therefore, along with the ACP in the device, it is possible to reject signals of other interference, for example, passive, in those cases when the latter are not dispersed along the co-ordinate coordinate. In addition, suppression of impulse noise, the sources of which are localized at certain elevation angles, is possible.

When assessing the effectiveness of the claimed device should consider the situation determined by the signals (3). The diagrams in Fig. 8 correspond to the case with two signals, one of which belongs to the ACP and the other to VO.

If there is no weighted signal processing during spectral analysis, then ΔВ ₁ ≈ -14 dB, and ΔВ ₂ ≈ -6 dB approximately corresponds to an octave decrease in the level of side lobes. Thus, for an ACW with a level of +60 dB relative to the receiver noise, ≈8 discrete ΔВ _{2 is} required so that the ACW level decreases to a threshold signal of ≈-3 dB for the signal at the input of the spectrum analyzer. It should be noted that in this case, a lot of (≥16) resolution elements along the angular coordinate should be implemented in the PAR. In addition, it should be noted that the symmetry of the spectrum and the greater likelihood of the jammer being located not only in the lower region of the fragile area worsen the situation with the observation of the HE signal, since the mutual overlap of the ACP spectra and the VO signal increases. Thus, it can be argued that at an ACW level of +60 dB, the absence of weighted signal processing and the number of PAR elements of order 16, it is impossible to observe the VO signals. When using the claimed device, there is the possibility of deep rejection of intense signals. Modeling of the notch process showed that a 60 dB suppression of the flying signal is achievable. In this case, VO signals with a signal-to-noise ratio of -3 dB are observed when using spectral analysis.

подавления сигналов не происходит. Это связано со свойствами режекторного фильтра 18, спектральное представление сигналов в котором производится лишь по низкочастотным опорным функциям (полиномам невысоких степеней). Сигналы ВО, отличающиеся по угловой координате от источника АШП, имеют иную более высокую частоту и не оставляют "следов" в спектре. Поэтому компенсации их не происходит и они наблюдаются на выходе режекторного фильтра и второго анализатора спектра.It can be shown that, in accordance with relations (4, 5), the suppression of ACP is determined by some frequency response. Figure 8 shows such a frequency response. We see that as the difference in signal frequencies increases, the degree of suppression decreases, but outside

signal suppression does not occur. This is due to the properties of the notch filter 18, the spectral representation of the signals in which is performed only by low-frequency reference functions (polynomials of low degrees). VO signals that differ in angular coordinate from the ACP source have a different higher frequency and do not leave “traces” in the spectrum. Therefore, they are not compensated and they are observed at the output of the notch filter and the second spectrum analyzer.

The use of weighting in spectral analysis improves, as you know, the situation with protection against ACP, because in this case, the relative level of the first side lobe decreases and in some cases, the rate of subsidence of the side lobes increases. However, the data [7 p. 434] show that, for example, the complex weight functions of Dolph-Chebyshev, Hamning and Taylor can only reduce the level of side lobes to a constant value of the order of (40-44) dB. Under these conditions, it is obvious that even for an interference of ≈50 dB this is not enough and signal detection is impossible. The use of the inventive device in these conditions compensates for the ACP and, even under these conditions, allows detecting the VO signals considered in the application.

Thus, taking into account that at present it is technically achievable to compensate signals whose level is 60 dB higher than the receiver noise level, it can be argued that it is possible to process VO signals in terms of detecting and measuring the frequency received in radars with headlamps from the same azimuthal direction from which the source of active noise interference is acting.

LITERATURE

1. Guide to radar, ed. M.Skolnika, t. 4. M .: Sov. Radio, 1979.

2. A.S. No. 107478 from 8.07.77 on the application No. 2.213.918 / 09 dated 10.01.77 (prototype).

3. Theoretical Foundations of Radar, ed. V.E.Dulevich, Vol. 2. M .: Sov. Radio, 1978.

4. Handbook of Radar, ed. M.Skolnik, vol. 3. M .: Sov. Radio, 1979.

5. Goncharov V.L. Theory of interpolation and approximation of functions, Gos. published technical and theoretical literature, M., 1954.

6. Kharkevich A.A. Spectra and analysis, State. published technical-theoretical literature, M., 1953.

7. Integrated circuits. Handbook Ed. Tarabrina B.V. M .: Radio and communications, 1984.

8. Kutyrkin Yu.M. and other foreign integrated circuits of wide application. Directory. M .: Energoatomizdat, 1984.

9. Rabiner L., Gould B. The theory of digital signal processing. M .: Mir, 1978.

10. Modern radar, per. from English, ed. Kobzareva Yu.B. M .: Sov. radio, 1969.

Claims (1)

A signal processing device in a radar with phased antenna arrays containing N antenna elements, N receiving channels connected to them, each of which consists of a series-connected microwave amplifier, mixer, matched filter, quadrature phase detector, electronic key and memory device, local oscillator the output of which is connected to the second inputs of the mixers of the receiving channels, a coherent local oscillator, the output of which is connected to the second inputs of the quadrature phase detectors of the receiving channels, switch, the inputs of which are connected to the outputs of the storage devices of the receiving channels, characterized in that, in order to provide protection from active noise interference received by the main lobe of the radiation pattern in the horizontal plane, as well as other types of high-intensity interference, forming correlated signal samples along the phased aperture antenna array, introduced in series connected quadrature analog-to-digital converter (ADC), the first quadrature spectrum analyzer, parameter meter signals, a threshold device and read-only memory (ROM), a quadrature delay device, a first quadrature multiplier, a quadrature notch filter, a second quadrature multiplier and a second quadrature spectrum analyzer, the output of which is connected to the second input of the signal parameter meter, the second output connected to the second the input of the ROM, as well as the delay device, and the first and second inputs of the quadrature ADC are connected respectively to the first and second outputs of the switch a, the first and second outputs of the ADC are connected respectively to the first and second inputs of the quadrature delay device, the first output of the ROM is connected to the heterodyne input of the first quadrature multiplier, and the second output of the ROM is connected to the heterodyne input of the second quadrature multiplier through the delay device, the first input of the clock pulses is connected to the clock inputs switch, quadrature ADC, quadrature delay device, ROM, quadrature notch filter and delay device, the second input of the clock is connected to the clock passages electronic key receiving channels, first and second quadrature spectrum analyzers and measuring signal parameters, the output of the threshold device and the third output measuring signal parameters are respectively first and second outputs of the device.

Images