RU2759117C1 - Method for nonlinear radar - Google Patents

Abstract

FIELD: radio engineering.

SUBSTANCE: invention relates to the field of radio engineering, in particular to methods and techniques for nonlinear radar, and can be used to search and detect objects with nonlinear electrical properties. The method is aimed at eliminating masking of weak echoes with strong ones and ensuring the reception of relatively weak echoes against a background of noise and interference. In the method, the number of partial radio pulses is selected from the condition Μ=2^k, where k=2, 3, …, and the echo signal, simultaneously with filtering by the matched filter, is filtered by the first additional filter, the first half of the impulse response of which is opposite to the first half of the impulse response of the matched filter, and the second half coincides with the second half of the impulse response of the matched filter, and the second additional filter with an impulse response, the first half of which is coincides with the second half of the impulse response of the first additional filter, and the second half coincides with the first half of the impulse response of the first additional filter. The signals from the outputs of the matched and the first additional filters are simultaneously summed and subtracted. The received signals are subjected to the first crossing procedure, the result of which is limited from below at the zero level. Similarly, the signals from the outputs of the matched and the second additional filters are simultaneously summed and subtracted, and the received signals are subjected to a second crossing procedure, the result of which is limited from below at the zero level. Signals limited after the first and second crossing procedures are subjected to the third crossing procedure.

EFFECT: compensation of the side peaks of the probing signal autocorrelation function and an increase in the noise immunity of the nonlinear radar.

1 cl, 11 dwg

Description

The invention relates to the field of radio engineering, in particular, to methods and techniques of nonlinear radar and can be used to search and detect objects with nonlinear electrical properties (ONES).

по сравнению с соответствующим кодом в зондирующем ФКМ-радиоимпульсе и, как следствие, к отсутствию эффекта увеличения дальности действия и улучшения разрешающей способности техники нелинейной радиолокации.An analogous method is a method of nonlinear radar, for example [1, P.156-162], based on the use of a specific effect of transforming the spectrum of the probing signal (ES) of the ONES, including the reception of echo signals from ONES at the second and third harmonics of the ES, processing and indication their levels for the operator to recognize the ONES. This is due to the fact that usually radar targets containing ONES with semiconductor components have a response signal level at the second harmonic that is 20-30 dB higher than at the third harmonic. For ONES of the contact type, as a rule, the inverse relation is fulfilled. The disadvantages of the analogue method include the short range and low range resolution of nonlinear radar technology, as well as the fact that the use of a phase-shift keying (PCM) radio pulse in the analogue method of processing (compression) of a phase-shift keyed (PCM) radio pulse will lead to a change in the intra-pulse phase keying code in the echo signal from ONES at the n-th harmonic of the ES

in comparison with the corresponding code in the probing FKM radio pulse and, as a consequence, to the absence of the effect of increasing the range and improving the resolution of nonlinear radar technology.

In addition, the analogous method does not provide for the elimination of side peaks of the autocorrelation function (ACF) of the PCM radio pulse in the case of its use and compression, as well as the necessary immunity of reception of relatively weak echo signals against the background of noise and interference.

, уменьшении в n раз значения начальной фазы каждого из Μ парциальных радиоимпульсов формируемого зондирующего ФКМ-радиоимпульса, где

- номер гармоники зондирующего сигнала, его излучении в зондируемую область пространства, приеме эхо-сигнала от объекта с нелинейными электрическими свойствами на частоте n-й гармоники ЗС, обработке эхо-сигнала в согласованном фильтре с импульсной характеристикой (ИХ), зеркальной по отношению к закону внутриимпульсной манипуляции фазы сформированного зондирующего ФКМ-радиоимпульса.The closest in technical essence and the achieved result to the claimed method of nonlinear radar (prototype for the alleged invention) is the method of nonlinear radar, given in the patent [2] and consists in the formation of a FKM radio pulse of a given duration by closing Μ> 1 partial radio pulses of the carrier frequency of the probing signal ƒ ₀ , the same amplitude u ₀ , the same duration τ ₀ for a limited number Ρ> 1 different possible values of the initial phase of oscillations ϕ _{i of} each of the partial radio pulses, where

, a decrease in n times the value of the initial phase of each of the Μ partial radio pulses of the formed probe PCM radio pulse, where

- the number of the harmonic of the probing signal, its emission into the probed area of space, the reception of an echo signal from an object with nonlinear electrical properties at the frequency of the n-th harmonic of the ES, processing of the echo signal in a matched filter with an impulse response (IM), mirror-like in relation to the law intrapulse manipulation of the phase of the formed probe PCM radio pulse.

The prototype method provides, in comparison with the analogue method, the processing of echo signals at the frequency of the n-th harmonic of the sounding signal, improves the range resolution and range of a nonlinear radar station (radar).

However, the prototype method does not allow to exclude the side peaks of the autocorrelation function of the FCM radio pulse, masking weaker signals reflected from objects against the background of stronger ones. The relative maximum level of these peaks depends on the class of the applied signal. The lowest level of side peaks is possessed by PCM signals based on Barker codes. For them, this level is 1 / Μ, where M≤13, which limits the possibility of using signals of this class, since the values of M >> 13 are required to ensure the required range of the nonlinear radar. In addition, the very presence of side peaks for any Μ always masks weaker echoes.

до

[3]. Для других классов сигналов этот уровень еще больше и может достигать 30-40% от основного пика. Поэтому проблема компенсации боковых пиков АКФ для любого класса сигналов и любого значения Μ актуальна.The most commonly used PCM signals based on Huffman codes have an ACF peak level equal to

before

[3]. For other classes of signals, this level is even higher and can reach 30-40% of the main peak. Therefore, the problem of compensating for the side peaks of the ACF for any class of signals and any value of Μ is relevant.

In addition to the above, the prototype method does not provide the necessary immunity of reception of relatively weak echo signals against the background of noise and interference. It is these signals in most cases that are echo signals of a nonlinear radar, since the power of the echo signal received at the n-th harmonic is proportional to the root of the 2 (n + 1) -th power from the range to ONES [4]. Noise and interference masks weak echoes at the nth harmonic of the carrier frequency.

The technical result of the invention is the elimination of the side peaks of the autocorrelation function of the PCM echo signal and the improvement of the noise immunity of the nonlinear radar under conditions of noise and interference.

The technical result is achieved by the fact that in the known prototype method, before the formation of the PCM signal, the number of partial radio pulses is selected from the condition M = 2 ^k , where k = 2, 3, ..., with a limited value of k, and the echo signal is simultaneously filtered with a matched filter filtered by the first additional filter, the first half of the impulse response of which is opposite to the first half of the impulse response of the matched filter, and the second half coincides with the second half of the impulse response of the matched filter, and the second additional filter with an impulse response, the first half of which coincides with the second half of the impulse response of the first additional filter , and the second half coincides with the first half of the impulse response of the first additional filter, the signals from the outputs of the matched and the first additional filters are simultaneously summed and subtracted, the summed and subtracted signals are subjected to the first procedure. of the section, the result of which is limited from below at the zero level, similarly, the signals from the outputs of the matched and the second additional filters are simultaneously summed and subtracted, the summed and subtracted signals are subjected to the second crossing procedure, the result of which is limited from below at the zero level, the signals limited after the first and second crossing procedures are subjected to the third crossing procedure.

The essence of the proposed method is as follows. The choice of the number of partial pulses from the condition M = 2 ^k , where k = 2, 3, ..., with a bounded value of k allows, when setting the law of intra-pulse manipulation of the phase of the formed PCM radio pulse for any value of k, to represent the radio pulse in the form of two halves with the number of partial pulses 2 ^k-1 in each, which makes it possible, when processing the echo signal (taking into account the newly introduced operations) to ensure the sameness and antiphase of the side peaks of the ACF at the output of the SF, available in the prototype, and the peaks of the cross-correlation functions (CCF) at the outputs of the first and second additional filters (DF).

This is implemented by selecting the impulse responses of additional filters so that the first half of the IR of the first DF coincides with the corresponding half of the IR SF, and the second half is opposite to the second half of the IR SF. In the second DF, the first half of the IR coincides with the second half of the IR of the first DF, and the second half coincides with the first half of the first DF. Further, the introduction of the operations of pairwise total-difference processing of the signals of the SF and the first DF and the SF and the second DF ensures the equality of the amplitudes of the antiphase side peaks of the SF and the first and second DF. Then, the introduction of the first and second intersection operations makes it possible to compensate for a part of the side peaks of the SF ACF in one case to the left, and in the other case to the right of the main peak of the SF ACF by using the properties of choosing a smaller one inherent in this operation, and the introduction of lower limiting operations at a zero level after intersection operation eliminates residual peaks of negative polarity. Finally, the introduction of the third operation of crossing limited signals leads to the complete elimination of the side peaks of the ACF SF.

Similarly to the elimination of the side peaks of the ACF SF, the claimed method compensates for noise and interference due to the newly introduced operations, providing an increase in noise immunity. Indeed, the organization of two-channel processing based on additional filters, and the sum-difference processing in each channel provides antiphase of noise and interference in additional filters relative to the matched filter. This makes it possible to implement their partial compensation in the channels using the first and second procedures of crossing and limiting from below. The third crossover procedure increases the degree of noise and interference compensation by combining the channel signals.

The difference in the processing of noise and interference in comparison with the processing of the side peaks of the ACF SF is due to the fact that it is possible to achieve complete correlation and antiphase of the side peaks of the ACF SF and CCF of additional filters and completely exclude the side peaks of the ACF. Due to the randomness of noise and interference, they cannot be completely eliminated, but in comparison with the prototype, as will be shown below, they are significantly weakened.

The proposed method of nonlinear radar is illustrated by graphic material.

Figure 1 shows: nonlinear radar, consisting of a reference generator - 1; devices for the formation of Μ partial impulses - 2; PCM signal forming devices - 3; transmitter - 4; transmitting antenna - 5; receiving antenna - 6; receiver - 7; matched filter - 8; the first additional filter - 9; the second additional filter - 10; the first adder - 11; the first block of subtraction - 12; the second adder - 13; the second block of subtraction - 14; the first block of intersection - 15; the second block of intersection - 16; the first stop from below at the zero level - 17; the second limiter from the bottom at the zero level - 18; the third block of intersection - 19; connected as shown in Fig. 1; object with nonlinear electrical properties - 20.

Figure 2 shows: 21 - signal at the output of the shaper Μ partial pulses 2; 22 - signal at the output of the device for forming the FCM signal 3; 23 - type of echo signal reflected from ONES; 24 - kind of signal from the output of the receiver 7.

Figure 3 shows the amplitude spectra: 25 - the probe signal from the output of the transmitter 4; 26 - echo signal reflected by ONES; 27 - the received signal at the output of the receiver 7.

Figure 4 shows the impulse response of the filters: 28 - IH SF 8; 29 - IH DF1 (block 9), 30 - IH DF2 (block 10).

In Fig. 5, the first three diagrams (22-24) repeat those already shown in Fig. 2. The difference from the plots in Fig. 2 lies in the shift of the echo signal 23 and the signal at the output of the receiver 7 relative to the origin of coordinates for a certain time delay due to the presence of a distance to the ONES. The following diagrams in figure 5 indicate: 31 - signal at the output of the SF 8; 32 - signal at the output of DF1 (block 9); 33 - signal at the output of the first adder 11; 34 - signal at the output of the first block of the difference 12; 35 - smoothed signal value at the output of the first limiter 17; 36 - signal at the output of DF2 (block 10); 37 - signal at the output of the second adder 13; 38 - signal at the output of the second block of the difference 14; 39 - smoothed signal value at the output of the second limiter 18; 40 - the envelope of the resulting autocorrelation function at the output of the claimed invention.

Figure 6 shows on a large scale: 41 - signal at the output of the prototype object; 42 - signal at the output of the claimed object.

In Fig. 7, analogous to Fig. 3, the amplitude spectra are shown: 25 - the probing signal from the output of the transmitter 4, 26 - the echo signal reflected from the ONES, 27 - the received signal at the output of the receiver 7. The difference from Fig. 3 lies in the presence of noise, hiding the useful signal.

The designations of the diagrams in FIG. 8 are similar to those in FIG. 4. The difference from Fig. 4 lies in the presence of noise hiding the useful signal.

Figure 9 shows the normalized envelopes of the output mixture of noise and useful signal: 43 - at the output of the SF of the prototype object (block 8); 44 - at the output of block 19 of the claimed object.

Figure 10 shows 22 - the probe signal of the transmitter, as well as the results of the processing of the echo signal and introduced to assess the noise immunity of interference acting against the background of noise: 45 - at the input of the receiver 7; 46 - at the output of the receiver 7; 47 - at the output of the matched filter 8; 48 - at the output of SF 8 (the result of processing in the prototype object); 49 - at the exit of the declared object (block 19). In this case, on plots 45-49, the signal and the introduced interference are indicated by symbols: I - echo signal from ONES; II - FKM impulse noise in the form of a 7-element Barker code; III - noise interference; IV - interference in the form of a short radio pulse; V - interference in the form of a long radio pulse.

Figure 11 shows the envelopes of the mixture of signal, noise and interference normalized in each case to the maximum level of interference: 50 - at the output of the SF (block 8) of the prototype object; 51 - at the output of the claimed object (block 19) plot 51. The signal and the introduced interference are designated by the same symbols as in Fig.10.

The inventive method is expressed in the following.

, и может быть построено, например, по известной схеме формирования D-кода, описанной в источнике [5]. Значение k ограничивается реализацией требуемой дальности действия РЛС, определяя степень накопления принимаемого эхо-сигнала в СФ, поскольку дальность пропорциональна

. Устройство формирования ФКМ-сигнала 3, выполненное, например, по известной схеме [6, с. 424-425, рис. 2.36], обеспечивает формирование ФКМ-радиоимпульса с уменьшенными в n раз значениями начальных фаз ϕ_i парциальных радиоимпульсов, поступивших с устройства 2, в соответствии с выражением

, где

- номер гармоники зондирующего сигнала.The reference generator 1 generates a signal representing the electromagnetic oscillations of the carrier frequency ЗС ƒ ₀ . Device 2 for the formation of Μ partial pulses in accordance with the expression M = 2 ^k , where k = 2, 3, ..., with a limited value of k in accordance with the selected code, provides the receipt of an FCM radio pulse in the composition of M = 2 ^k partial radio pulses of the carrier frequency, the same amplitude u ₀ , the same duration τ ₀ with a limited number of P> 1 different possible values of the initial phase of oscillations ϕ _{i of} each of the partial radio pulses, where

, and can be built, for example, according to the well-known D-code generation scheme described in the source [5]. The value of k is limited by the implementation of the required range of the radar, determining the degree of accumulation of the received echo signal in the SF, since the range is proportional to

... The device for the formation of the PCM signal 3, made, for example, according to the well-known scheme [6, p. 424-425, Fig. 2.36], provides the formation of a PCM radio pulse with n times reduced values of the initial phases ϕ _{i of the} partial radio pulses received from device 2, in accordance with the expression

, where

- harmonic number of the probing signal.

The transmitter 4 amplifies the PCM radio pulse of frequency ƒ ₀ and feeds it to the input of the transmitting antenna 5, with the help of which the formed ES is emitted into the given region of space.

, поскольку ЗС преобразуется нелинейным элементом в эхо-сигнал на частоте n-й гармоники с увеличенными в n раз начальными фазами парциальных радиоимпульсов. Приемник 7 усиливает сигналы, поступившие на его вход с выхода приемной антенны 6 и подает их на вход согласованного фильтра 8.The receiving antenna 6 serves to receive the echo signal scattered by ONES 20 at the frequency of the n-th harmonic of the ES. This signal is an FCM radio pulse of frequency nƒ ₀ , but already with the law of intra-pulse phase manipulation, corresponding to the code of length M = 2 ^k set in device 2 with a sequence of values of the initial phases

, since the ES is converted by a nonlinear element into an echo signal at the frequency of the n-th harmonic with the initial phases of partial radio pulses increased by a factor of n. The receiver 7 amplifies the signals received at its input from the output of the receiving antenna 6 and feeds them to the input of the matched filter 8.

The matched filter 8 is designed to compress the echo signal from the ONES, which is a PCM radio pulse at a frequency nƒ ₀ . To implement the compression, the impulse response of the SF is mirror-like with respect to the intra-impulse phase keying law set in device 2 according to the code of length M = 2 ^k adopted for operation.

Simultaneously (in parallel) with the SF, the echo signal is processed in the first 9 (DF1) and the second 10 (DF2) additional filters, built, for example, like the SF, according to the well-known scheme [7, P.436-437, Fig. 16.6], but with impulse characteristics different from SF. The first half of the impulse response DF1 is opposite to the first half of the impulse response of the SF, and the second half coincides with the second half of the IH SF. In the second additional filter, the first half of IH coincides with the second half of IH DF1, and the second half of IH DF2 coincides with the first half of IH DF1. Such a construction of the IR and the inclusion of additional filters in parallel with the LF ensures that signals at their outputs are obtained, which are cross-correlation functions with zero values of the peaks in the place of the maximum of the ACF of the SF and antiphase values of the peaks relative to the ACF in one case (for one DF) on the one hand relative to the maximum of the ACF, and in the other case (for another DF) on the other hand relative to the maximum of the ACF.

Further, the signals from the output of the SF are simultaneously fed to the first inputs of the first adder 11 and the first subtraction unit 12, the second adder 13, and the second subtractor unit 14. The adders and subtraction units can be performed according to a conventional amplifier circuit for two inputs or with direct and inverse inputs. the type described in [8, P.91, Fig. 2.23].

Signals from the output of DF1 are simultaneously supplied to the second inputs of the first adder 11 and the first block of subtraction 12, and signals from the output of DF2 are simultaneously supplied to the second inputs of the second adder 13 and the second block of subtraction 14.

After the sum-difference processing, the signals from the output of the first adder 11 are fed to the first input of the first block of intersection 15, to the second input of which signals are supplied from the output of the first block of subtraction 12. Then the signals from the output of the first block of intersection 15, passing the first limiter from below at the zero level 17 , arrive at the first input of the third block of intersection 19. Blocks 11, 12, 15, 17 form the first processing channel.

Similarly to the first channel, the second channel is organized as a part of blocks 13, 14, 16, 18. For this, after the sum-difference processing, the signals from the output of the second adder 13 are fed to the first input of the second block of intersection 16, to the second input of which signals are fed from the output of the second block of subtraction 14. Then the signals from the output of the second block of intersection 16, passing the second limiter from the bottom at the zero level 18, arrive at the second input of the third block of intersection 19.

Limiters from below at the zero level can be made according to a simple diode detector circuit [9, P.140, Fig. 5.12].

Intersection blocks can be implemented on the basis of adders, subtractors and modules for calculating the module [10, P.14, Fig. 1].

The task of the first processing channel is to minimize side peaks on one side (for example, to the left) of the main peak of the ACF SF and exclude these peaks on the other side (for example, to the right) of the main peak of the ACF SF.

A similar problem is solved by the second processing channel with the only difference that the sides of minimizing and eliminating peaks are reversed.

In this case, with the help of the third block of intersection 19, full compensation of the side peaks of the ACF SF is provided.

As for noise and interference, the organization of channel-by-channel processing based on the total difference transformation using the intersection operation and taking into account the previously indicated selection of the impulse characteristics of additional filters ensures the minimization of noise and interference and their significant decorrelation at the inputs of the third intersection block 19, which allows to a large extent compensate for them, improving noise immunity.

Let us explain the intersection operation used in the claimed method, the properties and structural implementation of which are given, for example, in [10, p. 13-17], which in the general case has the form

where x (t) and y (t) are arbitrary functions of time or signals.

This expression can be represented in a different form:

Both expressions are equivalent, but the physical meaning of the procedure is better understood from the last relation. It follows from it that the intersection procedure provides the choice of twice the lesser in absolute value of the two compared values (signals) with a sign equal to the product of the signs of these values. The use of this operation allows you to exclude the side peaks of the ACF and minimize the residual noise and interference.

The essence, performance and efficiency of the proposed method is illustrated by mathematical simulation. Let us take the operating frequency of the signal of the reference generator 1 ƒ ₀ equal to 5 MHz.

This signal is converted in the device for the formation of partial pulses 2 into a phase-shift keyed radio pulse in the composition Μ = 2 ^k , where in the general case k = 2, 3, ..., partial pulses with manipulations of the initial phases ϕ _i = {0, π) in accordance with the received code ... As code

the D-code is used [11, P.106-109]. Let us take k = 3, then M = 8, which makes it possible to obtain observable results in the simulation while preserving all the properties of the D-code. Then the sequence of information symbols corresponding to one of the 8-element D-code implementations will be: {0, 0, 0, 1, 1, 1, 0, 1}.

Let us associate the zero position of the code with the zero initial phase of the partial impulse, and the unit position with the phase equal to π. Then the sequence of values of the initial phases of partial radio pulses in the output signal of device 2 will have the form: {0, 0, 0, π, π, π, 0, π}. The form of a signal at a frequency of ƒ ₀ = 5 MHz with the obtained manipulation of the initial phases with the adopted in the model the duration of the partial pulse at the output of device 2 is shown in Fig. 2, plot 21, where the duration of the partial pulse is τ _o ≈ 0.7 μs.

. Будем исходить из необходимости приема эхо-сигнала от ОНЭС на второй гармонике несущей частоты ЗС. Тогда n=2, а последовательность значений начальных фаз парциальных радиоимпульсов в зондирующем ФКМ-радиоимпульсе приобретает вид: {0, 0, 0, π/2, π/2, π/2, 0, π/2}. Вид сигнала на выходе устройства 3 показан на фиг.2, эпюре 22.This signal enters the device for the formation of the PCM signal 3, which ensures the formation of the PCM radio pulse with the values of the initial phases ϕ _{i of the} partial radio pulses received from the device 2, reduced in the general case by a factor of n, in accordance with the expression

... We will proceed from the need to receive an echo signal from the ONES at the second harmonic of the ES carrier frequency. Then n = 2, and the sequence of values of the initial phases of the partial radio pulses in the probing PCM radio pulse takes the form: {0, 0, 0, π / 2, π / 2, π / 2, 0, π / 2}. The type of signal at the output of the device 3 is shown in Fig. 2, diagram 22.

Further, the signal is amplified in the transmitter 4 and emitted by the transmitting antenna 5 into a given area of space in the direction of the ONES.

In ONES, the emitted signal is converted into an echo signal at a frequency in the general case of the n-th harmonic of the ES with the initial phases of partial radio pulses increased by n times. In this case, the second harmonic is used, as the most powerful one in the echo signal.

Analysis [12] and modeling show that the work of a nonlinear element to generate a reflected signal at the second harmonic is quite satisfactorily approximated by the expression

where u _o (t) is the reflected echo signal;

u _s (t) is a probing signal (Fig. 2, plot 22).

The reflected echo is represented by plot 23 in FIG. 2. As follows from the simulation results, the second harmonic component dominates in the composition of this signal, although there are other components, including the constant component, and "jumps" of the phases of partial impulses are clearly manifested.

The reflected echo signal is received by the receiving antenna 6, amplified and subjected to band-pass filtering in the receiver 7, as in any receiver to reduce the level of noise and interference, and arrives simultaneously at the inputs of the matched filter 8, the first additional filter 9 and the second additional filter 10.

The form of the signal from the output of the receiver 7 is shown by the diagram 24 in Fig. 2. As follows from the simulation results, this signal is a PCM radio pulse at a frequency of 2ƒ ₀ (8 MHz) with phase manipulation of partial pulses {0, π), which repeat the phase manipulation at the output of device 2 (comparison of plots 21 and 24).

This is also evidenced by the representation of signals in the frequency domain in the form of amplitude spectra, shown in Fig. 3.

Figure 3 shows: 25 - spectrum of the ES from the output of the transmitter 4; 26 - spectrum of the echo signal reflected by ONES; 27 - spectrum of the received signal at the output of the receiver 7.

As can be seen from Fig. 3, the spectra of the probing and reflected signals differ both in the carrier frequency (two times) and in the shape. The signal spectrum from the receiver output is a typical amplitude spectrum of a PCM radio pulse with phase manipulation ϕ _i = {0, π} and the number of partial pulses Μ satisfying the condition Μ = 2 ^k , where k = 3. The width of the spectrum allocated by the receiver Δƒ is determined by the duration of the partial radio pulse and is expressed by the ratio Δƒ≈2 / τ _o .

Further, the signal from the output of the receiver 7 is subjected to simultaneous processing in SF - 8, DF1 - 9 and DF2 - 10 in accordance with the IH of these filters. The type of their filters is shown in Fig. 4.

Plot 28 is IH SF, plot 29 is IH DF1, and plot 30 is IH DF2. As follows from the comparison of the plots of the IR SF and the output signal of the receiver (plot 24 in Fig. 2), they are mirrored, therefore, the SF compresses the signal received at the second harmonic by a factor of Μ (in this case, Μ = 8) and an ACF is realized at its output. The impulse characteristics 28 and 29 of DF1 and DF2 were selected, as indicated earlier, and this is evident from their comparison with respect to the impulse characteristics of the SF and each other, so that due to the subsequent total-difference processing in the channels and the use of the petals of the ACF. Let us show this by illustrating the process of processing the received signal in the absence of noise, for a more accurate representation.

This process in the form of diagrams of signals at the output of the circuit elements (Fig. 1) is shown in Fig. 5 in real time with the parameters adopted in the simulation.

In Fig. 5, the first three diagrams repeat those already shown in Fig. 2, therefore, the previously adopted numbering is used for their designation: 22 - transmitter ES; 23 - echo signal scattered by ONES; 24 - reflected signal at the second harmonic at the output of the receiver. The difference from the diagrams in Fig. 2 lies in the shift of the echo signal relative to the origin of coordinates for a certain time delay t _s , due to the real presence of a certain distance between the nonlinear radar and ONES.

The corresponding time shift is acquired by all subsequent signals at the output of the circuit elements.

Diagram 31 is a signal at the output of SF 8, its instantaneous value u ₈ (t). This is a convolution of the signal from the output of the receiver and the IH SF or 8 times compressed input signal 24, which has the form of an ACF with the main lobe and side peaks that need to be compensated. The same diagram illustrates the signal at the output of the prototype object.

It should be noted that in order to implement the possibility of placing the signal diagrams on one figure, different amplitude coefficients were used. The compared signals are shown with the same coefficients.

Diagrams 31 ... 35 illustrate the operation of the first processing channel. Plot 32 is a signal at the output of DF1 u ₉ (t), which is a convolution of the signal from the output of the receiver 7 and IH DF1 and has the form VKF1. Diagram 33 is a signal at the output of the first adder 11 u ₁₁ (t) and is the coherent sum of the ACF from the output of the SF 8 block and the CCF from the output of the DF1 block: u ₁₁ (t) = u ₈ (t) + u ₉ (t). In turn, the plot 34 is the signal at the output of the first block of the difference 12 u ₁₂ (t) and is the coherent difference of the ACF from the output of the SF 8 block and the CCF1 from the output of the DF1 block: u ₁₂ (t) = u ₈ (t) -u ₉ (t).

The signal parameters adopted in the simulation and the scale used to illustrate the operation do not allow demonstrating the phase relationships of the ACF and CCF peaks, which are a consequence of the previously mentioned choice of the impulse characteristics DF1 and DF2 and are used to compensate for the side peaks of the ACF.

Further, the signals from the output of the first adder 11 and the first block of the difference 12 undergo an intersection procedure in the first block of intersection 15 in the form

where u ₁₅ (t) is the instantaneous voltage value at the output of the first block of intersection 15.

This signal is then subjected to bottom clipping at zero level in the first limiter 17. The smoothed value of the signal that has passed the limiter 17 and is the envelope of the crossing signal in the form of U ₁₇ (t) is shown in Fig. 5, plot 35. This signal is the result of the operation of the first channel and, as can be seen from the simulation results, it no longer has side ACF peaks to the right of the main peak, and only one peak to the left of the main peak remained uncompensated.

The problem of compensating for this peak is solved by the second processing channel, which operates similarly to the first with the only difference that instead of DF1, DF2 participates in the work.

Diagrams 36 ... 39 illustrate the work of the second processing channel. Diagram 36 is a signal at the output of DF2 u ₁₀ (t), which is a convolution of the signal from the output of the receiver 7 and IH DF2 and has the form VKF2. Diagram 37 is a signal at the output of the second adder 13 u ₁₃ (t) and is the coherent sum of the ACF from the output of the SF 8 block and the CCF2 from the output of the DF2 block: u ₁₃ (t) = u ₈ (t) + u ₁₀ (t). In turn, the plot 38 is a signal at the output of the second block of the difference 14 u ₁₄ (t) and is the coherent difference of the ACF from the output of the SF 8 block and the CCF2 from the output of the DF2 block: u ₁₄ (t) = u ₈ (t) -u ₁₀ (t).

The signals from the output of the second adder 13 and the second block of the difference 14 are subjected to the crossing procedure in the second block of the crossing 16 in the form

where u ₁₆ (t) is the instantaneous voltage value at the output of the second block of intersection 16.

Further, this signal is subjected to lower limiting at the zero level in the second limiter 18. The smoothed value of the signal that has passed the limiter 18 and represents the envelope of the crossing signal in the form of U ₁₈ (t) is shown in Fig. 5, plot 39. This signal is the result of the second channel and, as can be seen from the simulation results, the ACF has no side peaks to the left of the main peak, and only one peak to the right of the main peak remained uncompensated.

To fully compensate for the side peaks of the ACF, the signals from the outputs of the first processing channel U ₁₇ (t) and the second processing channel U ₁₈ (t) undergo an intersection procedure in the third intersection block 19 in the form

where U ₁₉ (t) is the envelope of the resulting ACF at the output of the claimed subject matter, which is shown in Fig. 5, diagram 40.

As follows from the simulation results, the side peaks of the ACF are fully compensated.

On a larger scale, a comparison of the results of signal processing in the prototype object and the claimed object is shown in Fig. 6, which presents the normalized envelopes of the ACF of the prototype object (plot 41) and at the output of the claimed object (plot 42). The main peaks of the ACF for both objects coincide, and all the side peaks of the ACF in the claimed object, in contrast to the prototype, are excluded.

Next, consider the process of noise compensation by the claimed object. To do this, we represent the reflected ONES echo signal u ₂₀ (t) as the sum of the useful signal u _c (t) and noise n (t), distributed according to the normal law with zero mean value and standard deviation (RMS) σ _w : u ₂₀ ( t) = u _c (t) + n (t). In this case, the ratio of the amplitude of the useful signal U _{c max} to the RMS deviation of the noise is assumed to be approximately equal to unity: U _{c max} / σ _w ≈1.

In the frequency domain, this is illustrated in Fig. 7, which shows the amplitude spectra: 25 - the probing signal from the output of the transmitter 4, 26 - the echo signal reflected by the ONES, 27 - the received signal at the output of the receiver 7. Since Fig. 7 in meaning repeats the amplitude spectra shown in Fig. 3, the numbering of the diagrams in Fig. 7 corresponds to the numbering of the diagrams in Fig. 3. The difference lies in the presence of noise hiding the useful signal, and this is noticeable in plots 25 and 26.

The noise compensation process is similar to that discussed and is illustrated by the signal diagrams at the outputs of the circuit elements shown in Fig. 8. These plots fully correspond in meaning and in terms of block outputs to the plots in Fig. 5, therefore the numbering of the plots is kept unchanged, and the noise compensation procedures are the same as when compensating for the lateral peaks of the ACF. In this case, at the beginning of processing (at the input of the receiver), the useful signal is practically hidden in the noise (Fig. 8, plot 23). Then, at the output of the receiver, due to band-pass filtering, the signal becomes noticeable, although not significantly (Fig. 8, plot 24). After the matched filter 8, which accumulates the useful signal, compressing it, the signal stands out noticeably against the background noise (Fig. 8, plot 31). This is also the case in the prototype object, but the noise level remains quite high. Further, in the claimed object, it is compensated together with the lateral peaks of the ACF, which is illustrated by plots 32 ... 39 in Fig. 8.

The result of processing the mixture of the useful signal and noise when the signal-to-noise ratio at the receiver input is approximately equal to unity is illustrated by the plot 40 in Fig. 8. As follows from the plot of the output signal of block 19, the side peaks of the ACF are fully compensated (as in the absence of noise) and noise is substantially suppressed. There are separate bursts of noise that have remained uncompensated.

An assessment of the degree of noise compensation in the claimed object in comparison with the prototype is shown in Fig. 9, which shows the normalized envelopes of the SF output mixture (block 8) of the prototype object (plot 43) and at the output of block 19 of the claimed object (plot 44). The main peaks of the ACF for both objects coincide, and all the side peaks of the ACF and most of the noise bursts in the claimed object, in contrast to the prototype, are suppressed.

A quantitative assessment of the degree of noise compensation by the claimed method was carried out by finding the ratio of the variance (power) of the noise at the output of the prototype object to the variance of the noise at the output of the claimed method. This variance ratio, obtained by averaging over the set of realizations, was at least 28.

This is the advantage of the proposed method in the noise immunity of receiving a relatively weak signal against a background of noise compared to the prototype.

, действующих на частоте и в полосе принимаемого от ОНЭС отраженного сигнала. В качестве помех используем: сигналоподобную коррелированную с эхо-сигналом ФКМ-импульсную помеху в виде 7-элементного кода Баркера с длительностью дискреты, равной длительности парциального импульса τ_o; шумовую помеху длительностью τ_ш≥Мτ_o; помеху в виде короткого радиоимпульса длительностью τ_и1<τ_o; помеху в виде длинного радиоимпульса длительностью τ_и2>Мτ_o. Тогда на входе приемника 7 действует смесь в виде u₂₀(t)=u_c(t)+n(t)+u_п(t), причем помехи не накладываются на полезный сигнал и раздельны во времени.To study the process of compensating for interference by the claimed object, we represent the echo signal reflected ONES u ₂₀ (t) as the sum of the useful signal u _c (t), noise n (t), distributed according to the normal law with zero mean value and RMS σ _w and a set of arbitrary typical interference

operating at the frequency and in the band of the reflected signal received from the ONES. As noise we use: signal-like correlated with the echo signal PCM-impulse noise in the form of a 7-element Barker code with a discrete duration equal to the duration of a partial pulse τ _o ; noise interference with duration τ _w ≥Mτ _o ; interference in the form of a short radio pulse with duration τ _{and 1} <τ _o ; interference in the form of a long radio pulse with duration τ _{and 2} > Mτ _o . Then, at the input of the receiver 7, a mixture acts in the form u ₂₀ (t) = u _c (t) + n (t) + u _p (t), and the interference is not superimposed on the useful signal and is separate in time.

Let us assume that the ratio of the amplitude of the useful signal U _{c max} to the RMS deviation of the noise is greater than one U _{c max} / σ _w ≈5, and the amplitudes of all received interference significantly exceed the amplitude of the useful signal U _{пi max} > U _{c max} .

The process of processing the input mixture of the useful signal, noise and interference is similar to the previously discussed processing of the useful signal or noise (Fig. 5, Fig. 8). Therefore, we will only explain the input signals and present the processing results shown in the form of voltage diagrams in Fig. 10, where the following are indicated: 22 - transmitter ES; 45 - signal at the input of the receiver 7 u ₂₀ (t), which includes an echo signal from ONES - I, FKM impulse interference in the form of a 7-element Barker code - II, noise interference - III, interference in the form of a short radio pulse - IV , interference in the form of a long radio pulse - V, acting against the background of noise; 46 - signals (echo and interference) at the output of the receiver 7, slightly changing their shape due to bandpass filtering; 47 - signals at the output of the matched filter 8 (useful signal in the form of ACF-I, interference - II, III, IV, V in the form of the corresponding CCF); 48 - the envelope of signals at the output of the SF, which is the result of processing in the prototype object; 49 - envelope of signals at the output of the claimed object (block 19).

As follows from the plot 48, at the output of the prototype object for given input signals and their parameters, a useful signal acts in the form of the main peak of the ACF and side peaks and all interference signals, the amplitudes of which, depending on the type of interference, are either comparable in level with the useful signal, or significantly it is exceeded.

At the same time, at the output of the claimed object (plot 49), there is a useful signal in the form of only the main peak of the ACF and, exceeding the useful signal, but significantly weakened correlated interference. As for other disturbances, there are minor residues.

A more clear assessment of the degree of interference compensation in the claimed object in comparison with the prototype is shown in Fig. 11, which shows the envelopes of the SF output mixture (block 8) of the prototype object (plot 50) normalized in each case to the maximum noise level and at the output of the claimed object ( block 19) plot 51. The main peaks of the ACF for both objects coincide, and all the side peaks of the ACF and most of the noise bursts in the claimed object, in contrast to the prototype, are suppressed.

A quantitative assessment of the degree of compensation used in the simulation of interference by the claimed method was carried out by finding the ratio of the normalized average dispersion (power) of interference at the output of the prototype object to the normalized average dispersion of interference at the output of the claimed method. This ratio was at least 10.

This is the advantage of the proposed method in the noise immunity of signal reception against the background of interference in comparison with the prototype.

Thus, the inventive method of nonlinear radar provides complete elimination of side peaks of the autocorrelation function of the PCM signal and improvement of the noise immunity of the nonlinear radar under conditions of noise and interference (in the cases considered, by 28 and 10 times, respectively).

Analysis of known technical solutions in the field of nonlinear radar shows that the claimed invention, due to essential features in the composition of the introduced operations and their sequence, determined the way to achieve the technical result, which consists in eliminating the lateral peaks of the autocorrelation function of the PCM signal and improving the noise immunity of the nonlinear radar in conditions of noise and interference, does not follow for a specialist explicitly from the prior art in this subject area and meets the requirement of "inventive step".

The applicant has not found an analogue characterized by features identical to all essential features of the claimed invention. The definition of the prototype, as the closest analogue in terms of the totality of features, made it possible to identify in the claimed object, significant in relation to the technical result, distinctive features, which allows us to consider the claimed invention satisfying the criterion of "inventive novelty".

The proposed technical solution is industrially applicable, since for its implementation can be used typical radio engineering units [8] and devices used in nonlinear radars, for example [1], as well as equipment and materials in the microwave range of a widespread technology [4, 6, 9, 13].

Sources of information:

1. Technical means and methods of information protection: Textbook for universities / Zaitsev A.P., Shelupanov Α.Α., Meshcheryakov R.V. and etc.; ed. A.P. Zaitsev and A.A. Shelupanova. - M .: OOO "Publishing House of Mechanical Engineering", 2009. - 508 p.

2. Likhachev VP, Usov Η.Α., Method of nonlinear radar // Patent of Russia №2382380. 2010. Bul. No. 5.

3. Noise-like signals in information transmission systems. Under. ed. prof. V.B. Pestryakov. M .: - Sov. radio, 1973 .-- 424 p.

4. Hstger R.О. Harmonic Radar Systems for Near-Ground In-Foliage Nonlinear Scatterers, IEEE Trans. on Aer. and El. Syst., 1976, vol. AES-12, no. 2, pp. 230-245.

5. Wilson R., Richter J. Generation and Performance of Quadraphase Welti Codes for Radar and Synchronization of Coherent and Differentially Coherent PSK, IEEE Trans, on Communications, 1979, vol. 27, no. 9, pp. 1296-1301.

6. Theory and technique of generation, emission and reception of radar signals: A textbook for students of the Academy / Ed. Yu.N. Syedysheva. - Kharkov: VIRTA, 1986 .-- 650 p.

7. Baskakov S.I. Radio circuits and signals: Textbook. for universities on specials. "Radiotekhnika", 3rd ed., Revised. and additional - M .: Higher school, 2000 .-- 462 p.

8. Alekseenko A.G. Application of precision analog microcircuits / A.G. Alekseenko, E.A. Colombet, G.I. Starodub. - 2nd ed., Rev. and additional - M .: Radio and communication, 1985 .-- 256 p.

9. Design of radar receiving devices / A.P. Golubkov, A.D. Dalmatov, A.P. Lukoshkin and others; Ed. M.A. Sokolov. - M., Higher. shk., 1984 .-- 335 p.

10. Universal multifunctional structural element of information processing systems / V.I. Gordienko, S.E. Dubrovsky, R.I. Ryumshin, D.V. Fenev // Radioelectronics. Izv. Universities, 1998. - No. 3. - S.13-17.

11. Varakin LE Communication systems with noise-like signals. - M .: Radio and communication, 1985 .-- 384 p.

12. Andreev V.S. Theory of nonlinear electrical circuits: Textbook for universities. - M .: Radio and communication, 1982 .-- 280 p.

13. Dulin V.N. Electronic and Quantum Microwave Devices: A Textbook for Higher Students. tech. study. institutions. 2nd ed., Rev. - M .: Energy, 1972 .-- 224 p.

Claims (1)

The method of nonlinear radar, which consists in the formation of a phase-manipulated radio pulse of a given duration by closing M> 1 partial radio pulses of the carrier frequency of the probing signal ƒ _{0 of the} same amplitude u ₀ , the same duration τ ₀ with a limited number of Ρ> 1 different possible values of the initial phase of oscillations ϕ _{i of} each of the partial radio pulses, where

, a decrease in the value of the initial phase of each of the Μ partial radio pulses of the generated probing phase-code-manipulated radio pulse by a factor of n, where

- the number of the harmonic of the probing signal, its emission into the probed area of space, the reception of an echo signal from an object with nonlinear electrical properties at the frequency of the n-th harmonic of the probe signal, filtering of the echo signal by a matched filter with an impulse response that is mirror-like with respect to the law of intra-pulse phase manipulation formed probing phase-keyed radio pulse, characterized in that the number of partial radio pulses is selected from the condition Μ = 2 ^k , where k = 2, 3, ..., and the echo signal is simultaneously filtered with a matched filter by the first additional filter, the first half of which half of the impulse response of the matched filter, and the second half coincides with the second half of the impulse response of the matched filter, and the second additional filter with impulse response, the first half of which coincides with the second half of the impulse response of the first th additional filter, and the second half coincides with the first half of the impulse response of the first additional filter, the signals from the outputs of the matched and the first additional filters are simultaneously summed and subtracted, the summed and subtracted signals are subjected to the first crossing procedure, the result of which is limited from below at a zero level, similarly signals with the outputs of the matched and the second additional filters are simultaneously summed and subtracted, the summed and subtracted signals are subjected to the second crossing procedure, the result of which is limited from below at a zero level, the signals limited after the first and second crossing procedures are subjected to the third crossing procedure.

Images