CN105137404A - Radar compression sampling method based on prepulse processing, and radar compression sampling system - Google Patents

Abstract

The invention proposes a radar compression sampling method and system based on pre-pulse inter-processing, belonging to the technical field of radar signal processing. The compressed sampling method includes: Step 1, using an analog pseudo-random sequence to mix the input analog signal to obtain a mixed frequency signal; Step 2, using a low-pass filter to filter the mixed frequency signal to obtain a filtered signal ; Step 3, use low-speed ADC to sample the input analog signal to obtain the sampled digital signal; Step 4, use the moving target indicator (MTI) and moving target detection (MTD) to perform pulse-to-pulse clutter suppression/coherent accumulation on the sampled digital signal Processing; step five, using a compression reconstruction algorithm to reconstruct the signal obtained after processing in step four. The compressed sampling method and system have the characteristics of low complexity, easy realization and the like.

Description

Radar compression sampling method and system based on pre-pulse processing

technical field

The invention relates to a radar compression sampling method and system based on pre-pulse inter-processing, and belongs to the technical field of radar signal processing.

Background technique

Traditional radar compression sampling system is based on analog-information converter (AnalogtoInformationConverter, AIC). First use an analog pseudo-random sequence to mix the analog signal, then pass the mixed signal through a low-pass filter, and then use a low-speed analog/digital converter (AnalogtoDigitalConverter, ADC) to undersample the filtered signal, and finally The product of the equivalent matrix of frequency mixing and low-pass filtering and the sparse matrix is used as the observation matrix, and the reconstruction algorithm of compressed sensing is used to restore the original signal from the under-sampled signal.

GuangmingShi proposed a waveform matching (WaveformMatch, WM) sparse dictionary for radar signal compression sampling. The WM sparse dictionary matrix can be regarded as a matrix composed of column vectors formed by a series of transmitted pulses with different time delays. It can be seen that this is very similar to the autocorrelation matrix of the transmitted signal. In fact, the compressed sampling and reconstruction of radar signals can be regarded as simultaneous sampling and pulse compression of the signal. Under the WM dictionary, the compressed and reconstructed target signal is the result of pulse compression of the radar signal.

In compressed sampling theory, the number of samples required for reliable recovery is determined by the sparsity and length of the reconstructed signal. When the length of the reconstructed signal is N and the degree of sparsity is K, the number of samples needed to recover the original signal reliably from the compressed sampling signal is: CKlog(N/K), where C is a positive empirical parameter. It can be seen that when the sparsity of the target signal is low, the minimum number of samples required for compressed sampling is almost proportional to the sparsity.

In the actual environment of radar, in addition to the echo of the target to be detected, the echo signal usually also contains a large number of clutter reflected by obstacles such as the ground. Clutter has the same waveform as the target echo, so it will also appear as a sparse target in the WM dictionary. These clutters will greatly increase the sparsity of the target signal, thereby seriously reducing the sampling efficiency of the radar compression sampling system. How to eliminate the impact of clutter on compressed sampling is one of the main difficulties to be solved in the practical application of radar compressed sampling.

The traditional AIC radar compression sampling system combines the two processing steps of pulse compression and target detection of the traditional radar system into compressed reconstruction, so the signal gain brought by clutter suppression and coherent accumulation cannot be used in target detection. Therefore, we need a new radar compression sampling method, which enables the system to have the gains brought by clutter suppression and coherent accumulation during compression and reconstruction, so that radar compression sampling technology can be used in multiple clutter/strong noise environments. get practical application.

Contents of the invention

In order to overcome the problem that the traditional radar compression sampling method and system cannot use the signal gain brought by clutter suppression and coherent accumulation, the technical solutions adopted are as follows:

The present invention proposes a radar compression sampling method and system based on pre-pulse processing, and the compression sampling method includes:

In step 1, an analog pseudo-random sequence is used to mix the input analog signal to obtain a mixed frequency signal;

Step 2, using a low-pass filter to filter the mixed frequency signal to obtain a filtered signal;

Step 3, using a low-speed ADC to sample the input analog signal to obtain a sampled digital signal;

Step 4, using the moving target indicator (MTI) and moving target detection (MTD) to perform inter-pulse clutter suppression/coherent accumulation processing on the sampled digital signal;

Step five, using a compression reconstruction algorithm to reconstruct the signal obtained after processing in step four.

Preferably, the input analog signal is a radar signal; the bandwidth of the mixing signal is the same as that of the input analog signal, and the spectrum energy is modulated within the bandwidth of the entire signal; the bandwidth of the low-pass filter is the same as that of the low-speed ADC The sampling frequency is the same;

Preferably, the moving target indicator (MTI) adopts a two-line delay cancellation method to eliminate static clutter in the echo signal; the clutter suppression/coherent accumulation process is performed on the pulse dimension of the analog signal Perform; the compression and reconstruction process is performed on the distance dimension of the analog signal.

Preferably, the clutter suppression/coherent accumulation process is aimed at different pulse echoes of the analog signal, and the different pulse echoes are jointly processed; the compression reconstruction process is aimed at a single pulse echo of the analog signal The pulse echo is used to process components at different time points in the single echo.

Preferably, the compressed sampling steps of the radar signal are as follows:

Step 1, the radar signal x(t) is mixed with an analog pseudo-random sequence p(t)={1,-1} to obtain the corresponding mixed signal x _p (t)=x(t)·p( t); the bandwidth of the pseudo-random sequence p(t) is the same as the bandwidth of the radar signal x(t);

Step 2, using a low-pass filter h(t) to filter the mixed frequency signal to obtain a filtered signal x _h (t)=x(t)*h(t); the low-pass filter h(t) The bandwidth is Fs=500MHz;

Step 3, use the low-speed ADC to sample the input analog signal to obtain the sampled digital signal y(m)=x _h (mΔt ₁ );

The sampling frequency of the low-speed ADC is the same as the bandwidth of the low-pass filter h(t), wherein Δt ₁ =1/Fs is the sampling interval of the low-speed ADC;

Step 4, use the moving target indicator (MTI) and moving target detection (MTD) to perform inter-pulse clutter suppression/coherent accumulation processing to obtain the signal:

r _il '(m)=y _l (m)-2y _l+1 (m)+y _l+2 (m)

${\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; l</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; l</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; W</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; l</span>}}$

Wherein, the number of coherent accumulation points of the moving target detection is L=3; the y _l (m) is the compressed sampling result of the lth pulse echo x _l (t) of the radar signal on the slow time dimension ; The r _il '(m) is the result of carrying out second-order anti-clutter suppression processing to y _l (m); r' _dl (m) is the result of coherent accumulation processing to y _l (m); W ^nl is a Fourier transform operator;

Step 5, using the compression reconstruction algorithm to compress and reconstruct the signal obtained in step 4 to obtain a reconstructed signal x(n), the waveform of the reconstructed signal x(n) is the same as the normal Nyquis The pulse compression results of the extra-sampled radar signal are the same; the reconstruction matrix Φ=H*P*F, wherein, P is a diagonal matrix composed of elements in p(t), and H is an analog low-pass filter h(t) The equivalent convolution matrix of F is the pulse compression matrix of the radar; the number of pulse compression points of the radar signal is I=64.

Preferably, the radar signal is a linear frequency modulation signal, and the pulse width of the radar signal is B=2GHz; the number of transmission points of the radar signal is N=64, the pulse transmission frequency is PRF=250MHz, and the carrier frequency is Fi=20GHz; the radar signal The number of targets contained in the signal is K=2, and the number of clutters is Kc=4.

Preferably, the moving target indicator (MTI) adopts a two-line delay cancellation method to eliminate static clutter in the echo signal; the clutter suppression/coherent accumulation processing is performed on the pulse dimension of the radar signal; The compression reconstruction process is performed on the range dimension of the radar signal. The clutter suppression/coherent accumulation processing is for different pulse echoes of the radar signal, and the different pulse echoes are jointly processed; the compression and reconstruction processing is for a single pulse echo of the radar signal , processing components at different time points in the single echo.

Preferably, a system for implementing the above method, the system includes a demodulator 1, a low-pass filter 2, a low-speed ADC3, a pulse train/parallel converter 4, a clutter suppression module 5, and a coherent accumulation module 6 And compression reconstruction module 7; The signal input port of described demodulator 1 is exactly described system signal input port, in order to receive analog input signal; The signal output port of described demodulator 1 and described low-pass filter 2 is connected to the signal input port; the signal output port of the low-pass filter 2 is connected to the signal input port of the low-speed ADC3; the signal output port of the low-speed ADC3 is connected to the signal input port of the pulse train/parallel converter 4 ; The signal output port of the pulse train/parallel converter 4 is connected to the signal input port a ₁ ... a _n of the clutter suppression module 5; the signal output port of the clutter suppression module 5 is connected to the coherent accumulation The signal input ports b ₁ ... b _n of the module 6 are connected in one-to-one correspondence; the signal output port of the coherent accumulation module 6 is connected to the signal input port c of the compression reconstruction module 7; the compression reconstruction module 7 The signal input port d is used to receive the sparse dictionary signal, and the reconstruction signal output port e of the compression reconstruction module 7 is the signal output port of the system;

The demodulator 1 is used for mixing the pseudo-random sequence and the input analog signal to obtain a mixed frequency signal; the low-pass filter 2 is used to filter out high-frequency interference signals in the mixed frequency signal; the low-speed ADC3 is used to sample the input analog signal to obtain a sampled digital signal; the clutter suppression module 5 is operated in conjunction with the coherent accumulation module 6, and is used for clutter suppression/coherent accumulation processing of the sampled digital signal; the compression reconstruction Module 7 is used to implement compression and reconstruction of the output signal after clutter suppression/coherent accumulation processing to obtain a reconstructed signal.

Preferably, the input analog signal in the above system is a radar signal.

Preferably, the compression sampling steps of the radar signal in the above system are:

Step 1, the radar signal x(t) is mixed with an analog pseudo-random sequence p(t)={1,-1} to obtain the corresponding mixed signal x _p (t)=x(t)·p( t); the bandwidth of the pseudo-random sequence p(t) is the same as the bandwidth of the radar signal x(t);

Step 2, using a low-pass filter h(t) to filter the mixed frequency signal to obtain a filtered signal x _h (t)=x(t)*h(t); the low-pass filter h(t) The bandwidth is Fs=500MHz;

Step 3, use the low-speed ADC to sample the input analog signal to obtain the sampled digital signal y(m)=x _h (mΔt ₁ );

The sampling frequency of the low-speed ADC is the same as the bandwidth of the low-pass filter h(t), wherein Δt ₁ =1/Fs is the sampling interval of the low-speed ADC;

Step 4, use the moving target indicator (MTI) and moving target detection (MTD) to perform inter-pulse clutter suppression/coherent accumulation processing to obtain the signal:

r _il '(m)=y _l (m)-2y _l+1 (m)+y _l+2 (m)

${\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; l</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; l</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; W</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; l</span>}}$

Wherein, the number of coherent accumulation points of the moving target detection is L=3; the y _l (m) is the compressed sampling result of the lth pulse echo x _l (t) of the radar signal on the slow time dimension ; The r _il '(m) is the result of carrying out second-order anti-clutter suppression processing to y _l (m); r' _dl (m) is the result of coherent accumulation processing to y _l (m); W ^nl is a Fourier transform operator;

Step 5, using the compression reconstruction algorithm to compress and reconstruct the signal obtained in step 4 to obtain a reconstructed signal x(n), the waveform of the reconstructed signal x(n) is the same as the normal Nyquis The pulse compression results of the extra-sampled radar signal are the same; the reconstruction matrix Φ=H*P*F, wherein, P is a diagonal matrix composed of elements in p(t), and H is an analog low-pass filter h(t) The equivalent convolution matrix of F is the pulse compression matrix of the radar; the number of pulse compression points of the radar signal is I=64.

Preferably, the radar signal is a linear frequency modulation signal, and the pulse width of the radar signal is B=2GHz; the number of transmission points of the radar signal is N=64, the pulse transmission frequency is PRF=250MHz, and the carrier frequency is Fi=20GHz; the radar signal The number of targets contained in the signal is K=2, and the number of clutters is Kc=4.

Preferably, the moving target indicator (MTI) adopts a two-line delay cancellation method to eliminate static clutter in the echo signal; the clutter suppression/coherent accumulation processing is performed on the pulse dimension of the radar signal; The compression and reconstruction process is performed on the distance dimension of the radar signal; the clutter suppression/coherent accumulation process is aimed at different pulse echoes of the radar signal, and the different pulse echoes are jointly processed; The compression and reconstruction process processes components at different time points in the single pulse echo of the radar signal.

Preferably, the radar compression sampling system includes a demodulator 1, a low-pass filter 2, a low-speed ADC3, a burst/parallel converter 4, a clutter suppression module 5, a coherent accumulation module 6 and a compression reconstruction module 7; The signal input port of the demodulator 1 is the system signal input port for receiving an analog input signal; the signal output port of the demodulator 1 is connected to the signal input port of the low-pass filter 2; The signal output port of the low-pass filter 2 is connected with the signal input port of the low-speed ADC3; the signal output port of the low-speed ADC3 is connected with the signal input port of the pulse train/parallel converter 4; the pulse train/parallel The signal output port of the converter 4 is connected to the signal input ports a ₁ ... a _n of the clutter suppression module 5; the signal output port of the clutter suppression module 5 is connected to the signal input port b of the coherent accumulation module 6 ₁ ... b _n are connected in one-to-one correspondence; the signal output port of the coherent accumulation module 6 is connected to the signal input port c of the compression reconstruction module 7; the signal input port d of the compression reconstruction module 7 is used to receive Sparse dictionary signal, the reconstruction signal output port e of the compression reconstruction module 7 is the signal output port of the system;

The demodulator 1 is used for mixing the pseudo-random sequence and the input radar signal to obtain a mixed frequency signal; the low-pass filter 2 is used to filter out high-frequency interference signals in the mixed frequency signal; the low-speed ADC3 is used to sample the input radar signal to obtain a sampled digital signal; the clutter suppression module 5 is operated in conjunction with the coherent accumulation module 6 for clutter suppression/coherent accumulation processing of the sampled digital signal; the compression reconstruction Module 7 is used to realize the compression and reconstruction of the output signal after clutter suppression/coherent accumulation processing, and obtain the reconstructed signal;

The above modules are running, the specific steps are as follows:

Step 1, the radar signal x(t) is mixed with an analog pseudo-random sequence p(t)={1,-1} to obtain the corresponding mixed signal x _p (t)=x(t)·p( t); the bandwidth of the pseudo-random sequence p(t) is the same as the bandwidth of the radar signal x(t);

Step 2, using a low-pass filter h(t) to filter the mixed frequency signal to obtain a filtered signal x _h (t)=x(t)*h(t); the low-pass filter h(t) The bandwidth is Fs=500MHz;

Step 3, use the low-speed ADC to sample the input analog signal to obtain the sampled digital signal y(m)=x _h (mΔt ₁ );

The sampling frequency of the low-speed ADC is the same as the bandwidth of the low-pass filter h(t), wherein Δt ₁ =1/Fs is the sampling interval of the low-speed ADC;

Step 4, use the moving target indicator (MTI) and moving target detection (MTD) to perform inter-pulse clutter suppression/coherent accumulation processing to obtain the signal:

r _il '(m)=y _l (m)-2y _l+1 (m)+y _l+2 (m)

${\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; l</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; l</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; W</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; l</span>}}$

Wherein, the number of coherent accumulation points of the moving target detection is L=3; the y _l (m) is the compressed sampling result of the lth pulse echo x _l (t) of the radar signal on the slow time dimension ; The r _il '(m) is the result of carrying out second-order anti-clutter suppression processing to y _l (m); r' _dl (m) is the result of coherent accumulation processing to y _l (m); W ^nl is a Fourier transform operator;

Step 5, using the compression reconstruction algorithm to compress and reconstruct the signal obtained in step 4 to obtain a reconstructed signal x(n), the waveform of the reconstructed signal x(n) is the same as the normal Nyquis The pulse compression results of the extra-sampled radar signal are the same; the reconstruction matrix Φ=H*P*F, wherein, P is a diagonal matrix composed of elements in p(t), and H is an analog low-pass filter h(t) The equivalent convolution matrix of F is the pulse compression matrix of the radar; the number of pulse compression points of the radar signal is I=64.

Beneficial effects of the present invention:

1 The radar compression sampling method and system based on pre-pulse processing proposed by the present invention can well suppress the static clutter components in radar compression sampling, and provide the signal-to-noise ratio gain brought by coherent accumulation, so that the radar signal The probability of compression reconstruction in the environment of clutter and strong noise has been greatly improved.

2 The radar compression sampling method and system based on inter-pulse processing proposed by the present invention reduces the sparsity of the reconstructed signal and reduces the number of sampling points required for reliable reconstruction.

3 The radar compression sampling technology based on pre-pulse processing proposed by the present invention has the advantages of simple system implementation and low complexity. Since both clutter suppression and coherent accumulation technologies have mature theories and algorithms, the system has low complexity and is easy to implement.

Description of drawings

Fig. 1 is the specific implementation block diagram of the radar compression sampling method and system based on pre-pulse processing in the present invention

Fig. 2 is the structural representation of moving target indicator (MTI) bilinear canceller of the present invention;

Fig. 3 is a schematic diagram of moving target detection (MTD) clutter suppression/coherent accumulation;

Fig. 4 is a schematic diagram of the distance dimension and the pulse dimension of the radar echo signal;

Figure 5 is a comparison diagram of the reconstruction results of the pulse compression signal after clutter suppression/coherent accumulation processing and the radar compressed sampling signal after clutter suppression/coherent accumulation processing;

(a, MTI simulation results; b, MTD simulation results)

Figure 6 is a performance comparison diagram of the radar compression sampling method and system based on pre-pulse inter-processing and the reconstruction method of the traditional AIC system in the clutter environment;

Fig. 7 is a comparison chart of reconstruction performance between the radar compression sampling method and system based on pre-pulse inter-processing and the traditional AIC system in a noisy environment.

Detailed ways

The present invention will be further described below in conjunction with specific examples, but the present invention is not limited by the examples.

In conjunction with Fig. 1 to Fig. 4, the radar compression sampling method and system based on pre-pulse processing in the present invention are described in detail:

The compressed sampling steps of the radar signal are as follows:

Step 1, the radar signal x(t) is mixed with an analog pseudo-random sequence p(t)={1,-1} to obtain the corresponding mixed signal x _p (t)=x(t)·p( t); the bandwidth of the pseudo-random sequence p (t) is identical to the bandwidth of the radar signal x (t); the pulse width of the radar signal is B=2GHz; the number of sending points of the radar signal is N=64, and the pulse transmission frequency PRF=250MHz, carrier frequency Fi=20GHz; the number of targets contained in the radar signal is K=2, and the number of clutter is Kc=4.

The frequency mixing step spreads the energy of all frequencies of the input analog signal x(t) to the entire bandwidth of x(t), so as to prevent the loss of frequency information caused by undersampling, and at the same time provide the necessary randomness of the sampling matrix for sparse reconstruction;

Step 2, using a low-pass filter h(t) to filter the mixed frequency signal to obtain a filtered signal x _h (t)=x(t)*h(t); the low-pass filter h(t) The bandwidth is Fs=500MHz;

Step 3, use the low-speed ADC to sample the input analog signal to obtain the sampled digital signal y(m)=x _h (mΔt ₁ );

The sampling frequency of the low-speed ADC is the same as the bandwidth of the low-pass filter h(t), wherein Δt ₁ =1/Fs is the sampling interval of the low-speed ADC;

Step 4, perform inter-pulse clutter suppression/coherent accumulation processing on the sampled digital signal with two-line delay cancellation MTI and MTD with coherent accumulation points L=32:

r _il '(m)=y _l (m)-2y _l+1 (m)+y _l+2 (m)

${\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; l</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; l</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; W</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; l</span>}}$

Wherein, the number of coherent accumulation points of the moving target detection is L=3; the y _l (m) is the compressed sampling result of the lth pulse echo x _l (t) of the radar signal on the slow time dimension ; The r _il '(m) is the result of carrying out second-order anti-clutter suppression processing to y _l (m); r' _dl (m) is the result of coherent accumulation processing to y _l (m); W ^nl is a Fourier transform operator;

Step 5, using the compression reconstruction algorithm to compress and reconstruct the signal obtained in step 4 to obtain a reconstructed signal x(n), the waveform of the reconstructed signal x(n) is the same as the normal Nyquis The pulse compression results of the extra-sampled radar signal are the same; the reconstruction matrix Φ=H*P*F, wherein, P is a diagonal matrix composed of elements in p(t), and H is an analog low-pass filter h(t) The equivalent convolution matrix of F is the pulse compression matrix of the radar; the number of pulse compression points of the radar signal is I=64.

A system for realizing the above method, the system includes a demodulator 1, a low-pass filter 2, a low-speed ADC3, a pulse train/parallel converter 4, a clutter suppression module 5, a coherent accumulation module 6 and a compression heavy Constructing module 7; The signal input port of described demodulator 1 is exactly described system signal input port, in order to receive analog input signal; The signal output port of described demodulator 1 and the signal of described low-pass filter 2 The input port is connected; the signal output port of the low-pass filter 2 is connected with the signal input port of the low-speed ADC3; the signal output port of the low-speed ADC3 is connected with the signal input port of the pulse train/parallel converter 4; the The signal output port of the burst/parallel converter 4 is connected to the signal input ports a ₁ ... a _n of the clutter suppression module 5; the signal output port of the clutter suppression module 5 is connected to the coherent accumulation module 6 The signal input ports b ₁ ... b _n are connected in one-to-one correspondence; the signal output port of the coherent accumulation module 6 is connected to the signal input port c of the compression reconstruction module 7; the signal input port of the compression reconstruction module 7 D is used for receiving the sparse dictionary signal, and the reconstruction signal output port e of the compression reconstruction module 7 is the signal output port of the system;

The demodulator 1 is used for mixing the pseudo-random sequence and the input analog signal to obtain a mixed frequency signal; the low-pass filter 2 is used to filter out high-frequency interference signals in the mixed frequency signal; the low-speed ADC3 is used to sample the input analog signal to obtain a sampled digital signal; the clutter suppression module 5 is operated in conjunction with the coherent accumulation module 6, and is used for clutter suppression/coherent accumulation processing of the sampled digital signal; the compression reconstruction Module 7 is used to implement compression and reconstruction of the output signal after clutter suppression/coherent accumulation processing to obtain a reconstructed signal.

The radar compression sampling system includes a demodulator 1, a low-pass filter 2, a low-speed ADC3, a pulse train/parallel converter 4, a clutter suppression module 5, a coherent accumulation module 6 and a compression reconstruction module 7; The signal input port of the modulator 1 is the system signal input port for receiving the analog input signal; the signal output port of the demodulator 1 is connected with the signal input port of the low-pass filter 2; The signal output port of the pass filter 2 is connected with the signal input port of the low-speed ADC3; the signal output port of the low-speed ADC3 is connected with the signal input port of the pulse series/parallel converter 4; the pulse series/parallel converter 4 The signal output port of the clutter suppression module 5 is connected to the signal input port a ₁ ... a _n ; the signal output port of the clutter suppression module 5 is connected to the signal input port b ₁ ... b of the coherent accumulation module 6 _n is connected in one-to-one correspondence; the signal output port of the coherent accumulation module 6 is connected with the signal input port c of the compression reconstruction module 7; the signal input port d of the compression reconstruction module 7 is used to receive the sparse dictionary signal , the reconstruction signal output port e of the compression reconstruction module 7 is the signal output port of the system;

The demodulator 1 is used for mixing the pseudo-random sequence and the input radar signal to obtain a mixed frequency signal; the low-pass filter 2 is used to filter out high-frequency interference signals in the mixed frequency signal; the low-speed ADC3 is used to sample the input radar signal to obtain a sampled digital signal; the clutter suppression module 5 is operated in conjunction with the coherent accumulation module 6 for clutter suppression/coherent accumulation processing of the sampled digital signal; the compression reconstruction Module 7 is used to realize the compression and reconstruction of the output signal after clutter suppression/coherent accumulation processing, and obtain the reconstructed signal;

In conjunction with Fig. 2 to Fig. 4, the theoretical basis of the radar compression sampling method based on pre-pulse processing proposed by the present invention is described:

The radar system uses inter-pulse processing such as clutter suppression based on Moving Target Indicator (MTI) and coherent accumulation based on Moving Target Detection (MTD) to combat clutter and noise in radar signals. MTI radar uses the Doppler frequency shift of the moving target echo to distinguish the moving target and stationary clutter in the echo signal. In pulsed radar systems, this Doppler shift appears as a phase change in the echo signal between successively returning radar pulses. Taking advantage of this phase change, the MTI radar uses a delay canceller to remove stationary clutter from the echo signal and preserve some of the energy of the moving target:

r _il (n)＝x _l (n)-2x _l+1 (n)+x _l+2 (n)

After passing through the delay canceller, the static clutter with the same phase in different pulses of the radar echo signal is completely eliminated, while the moving targets with different phases retain part of their energy. Through this cancellation, the MTI system can eliminate the static clutter in the echo signal, thereby effectively reducing the sparsity of the compressed sampling reconstructed signal and improving the sampling efficiency of the radar compressed sampling system.

When the compressed sampling technique works in a noisy environment, it is required that the detected sparse peaks have obvious amplitude differences compared with the noise components. Conventional radar compression sampling systems rely on pulse compression for signal-to-noise ratio gain on the target signal to meet this requirement. However, when the signal-to-noise ratio of the radar echo signal is low, simple pulse compression often cannot meet the signal-to-noise ratio requirements of signal compression and reconstruction. At this point, the target signal will be overwhelmed by noise, thus losing the sparsity required for compressive sensing.

Traditional radar signal processing usually utilizes coherent accumulation technique to provide more SNR gain for radar signal. That is to accumulate multiple echo pulses of the radar signal, and obtain more reflected energy by increasing the irradiation time under a certain irradiation power. Currently the most widely used coherent accumulation technique is MTD technique. MTD is implemented using a Doppler filter bank. The passband of each Doppler filter covers a certain frequency range, and the output of each filter can determine whether there is a moving target in the frequency range, and can measure the speed of the moving target. MTD improves the frequency characteristics of the MTI filter, making it closer to the optimal (matched) linear filter, thereby increasing the improvement factor. When the Doppler frequency shift of the target is close to the central frequency band of the Doppler filter, the weighting factor of the target can eliminate the phase difference between each echo pulse, so that the signal energy in each pulse can be superimposed in phase, thereby obtaining the signal-to-noise ratio gain. In practical engineering applications, the Doppler filter bank of MTD can be realized with an FFT:

${\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; W</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; l</span>}}$

In radar signal processing, pulse compression and reconstruction and clutter suppression/coherent accumulation are respectively carried out in two dimensions (range dimension and pulse dimension). That is, pulse compression and reconstruction are aimed at a single pulse echo, and the components at different time points in the echo are processed. The clutter suppression/coherent accumulation is for joint processing of different pulse echoes. Let x _l (t) be the lth pulse echo of the radar signal in the slow time dimension, and y _l (m) be the compressed sampling result of x _l (t). According to the AIC sampling system mentioned above, y _l The relationship between (m) and x _l (t) is as follows:

${\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>$

Let r _il (m) be the result of clutter suppression processing on x _l (t), r _il ′(m) is the result of second-order cancellation clutter suppression processing on y _l (m), x _l (τ ) is the representation of x _l (t) in the cumulative formula, and we get:

$\begin{array}{c}{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; l</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span>\\ <span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 2</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>\\ <span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>\end{array}$

It can be seen from the above formula that the reconstruction result of the radar compressed sampling signal after clutter suppression processing is the same as the pulse compression signal processed by clutter suppression.

Similarly, let r _dl (m) be the result of L-point coherent accumulation processing on x _l (t), and r′ _dl (m) be the result of coherent accumulation processing on y _l (m), we can get:

$\begin{array}{c}{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; l</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; W</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; l</span>}}\\ <span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; W</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; l</span>}}\underset{\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>\\ <span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>\underset{\mathrm{<span\; class="notranslate">\; l</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; L</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>{\mathrm{<span\; class="notranslate">\; W</span>}}^{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; l</span>}}\\ <span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0</span>}}{\overset{\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; d</span>}\mathrm{<span\; class="notranslate">\; l</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; h</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; \&tau;</span>}<span\; class="notranslate">\; )</span>\end{array}$

This result is similar to the previous clutter suppression process. It can be seen that the reconstruction result of the radar compressed sampling signal after coherent accumulation processing is the same as that of the pulse compression signal processed by coherent accumulation.

In conjunction with Fig. 5 to Fig. 7, the effectiveness of the conversion method and system described in this embodiment is illustrated, and the specific simulation examples and analysis are as follows:

Fig. 5 is a comparison diagram of the reconstruction result of the pulse compression signal after clutter suppression/coherent accumulation processing and the radar compressed sampling signal after clutter suppression/coherent accumulation processing. The analog signal signal used in the simulation is a radar echo signal containing 2 moving targets and 4 stationary clutters. The pulse compression signal of this signal and the result of compressed sampling are separately clutter suppression and clutter suppression/phase After parameter accumulation processing, the waveform of the pulse compression signal is compared with the reconstruction result of the compressed sampling signal.

Wherein, the asterisk indicates the waveform of the pulse compression signal, and the circle indicates the reconstruction result of the compressed sampling signal.

The simulation results prove that the clutter suppression/coherent accumulation processing directly on the radar compressed sampling can suppress the clutter in the signal, increase the reconstructed signal-to-noise ratio of the signal, and reduce the sparsity. It can be seen from the simulation results that after the reconstruction of the original signal containing 2 moving targets and 4 stationary clutters, only the energy of 2 moving targets remains in the reconstruction result after the clutter suppression processing, and the energy of the stationary The energy of the noise components is almost completely eliminated. After the signal is further processed by coherent accumulation, the energy in the L echo pulses is accumulated together through coherent accumulation, which not only obtains an L-fold signal-to-noise ratio gain, but also reduces the sparsity of the reconstructed signal by an L-fold .

Fig. 6 is the reconstruction probability curves of the traditional AIC radar compressed sampling system and the radar compressed sampling system based on pre-clutter suppression/coherent accumulation proposed by the present invention under different under-sampling multiples in a clutter environment. It can be seen from Fig. 7 that the reconstruction performance of the proposed system in the present invention is much higher than that of the traditional AIC system in the clutter environment. At the same time, it should also be noted that in the actual environment, the amount of clutter contained in the radar echo signal is usually far more than the amount set in the simulation. Under such circumstances, the performance gap between the system proposed in this paper and the AIC system will be further improved.

Fig. 7 is when the sampling frequency is that the sampling frequency is Fs=500MHz, the traditional AIC radar compression sampling system and the radar compression sampling system based on pre-clutter suppression/coherent accumulation proposed by the present invention are reconstructed under different input signal-to-noise ratios probability curve. It can be seen from Fig. 7 that the reconstruction performance of the proposed method in the present invention is much higher than that of the traditional AIC system in a low SNR environment.

Although the present invention has been disclosed above with preferred embodiments, it is not intended to limit the present invention. Any person familiar with this technology can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore The scope of protection of the present invention should be defined by the claims.

Claims (12)

1. A radar compression sampling method and system based on pre-pulse processing, it is characterized in that, the compression sampling method comprises: In step 1, an analog pseudo-random sequence is used to mix the input analog signal to obtain a mixed frequency signal; Step 2, using a low-pass filter to filter the mixed frequency signal to obtain a filtered signal; Step 3, using a low-speed ADC to sample the input analog signal to obtain a sampled digital signal; Step 4, using the moving target indicator (MTI) and moving target detection (MTD) to perform inter-pulse clutter suppression/coherent accumulation processing on the sampled digital signal; Step five, using a compression reconstruction algorithm to reconstruct the signal obtained after processing in step four. 2. The method according to claim 1, wherein the input analog signal is a radar signal; the bandwidth of the mixed frequency signal is the same as the input analog signal bandwidth; the bandwidth of the low-pass filter is the same as that of the low-speed ADC The sampling frequency is the same. 3. The method according to claim 1, characterized in that, the moving target indicator (MTI) adopts a two-wire delay cancellation mode to eliminate static clutter in the echo signal; the clutter suppression/coherence The accumulation process is performed on the pulse dimension of the analog signal; the compression and reconstruction process is performed on the distance dimension of the analog signal; wave, performing joint processing on the different pulse echoes; the compression and reconstruction processing is aimed at a single pulse echo of the analog signal, and processing components at different time points in the single echo. 4. according to the method described in claim 2, it is characterized in that, the compressed sampling step of described radar signal is as follows: Step 1, the radar signal x(t) is mixed with an analog pseudo-random sequence p(t)={1,-1} to obtain the corresponding mixed signal x _p (t)=x(t)·p( t); the bandwidth of the pseudo-random sequence p(t) is the same as the bandwidth of the radar signal x(t); Step 2, using a low-pass filter h(t) to filter the mixed frequency signal to obtain a filtered signal x _h (t)=x(t)*h(t); the low-pass filter h(t) The bandwidth is Fs=500MHz; Step 3, use the low-speed ADC to sample the input analog signal to obtain the sampled digital signal y(m)=x _h (mΔt ₁ ). The sampling frequency of the low-speed ADC is the same as the bandwidth of the low-pass filter h(t), wherein, Δt ₁ =1/Fs is the low-speed ADC sampling interval; Step 4, use the moving target indicator (MTI) and moving target detection (MTD) to perform inter-pulse clutter suppression/coherent accumulation processing to obtain the signal: r′ _il (m)=y _l (m)-2y _l+1 (m)+y _l+2 (m) Wherein, the number of coherent accumulation points of the moving target detection is L=3; wherein, y _l (m) is the compressed sampling result of the lth pulse echo x _l (t) of the radar signal on the slow time dimension ; The r' _il (m) is the result of carrying out second-order cancellation clutter suppression processing to y _l (m); r' _dl (m) is the result of carrying out coherent accumulation processing to y _l (m), W ^nl is a Fourier transform operator; Step 5, using the compression reconstruction algorithm to perform compression reconstruction processing on the obtained signal processed in the step 4 to obtain the reconstruction signal x(n), the reconstruction matrix Φ=H*P*F, where P is p The diagonal matrix that the element in (t) forms, H is the equivalent convolution matrix of analog low-pass filter h (t), and F is the pulse compression matrix of radar; Described radar signal pulse compression point number is I=64. 5. according to the method described in claim 5, it is characterized in that, described radar signal is chirp signal, and radar signal pulse width is B=2GHz; Described radar signal transmission point is N=64, and pulse emission frequency is PRF= 250MHz, the carrier frequency is Fi=20GHz; the number of targets contained in the radar signal is K=2, and the number of clutter is Kc=4. 6. according to the method described in claim 5, it is characterized in that, described moving target indicator (MTI) adopts two-wire delay cancellation mode to eliminate the stationary clutter in the echo signal; The parameter accumulation process is carried out on the pulse dimension of the radar signal; the compression and reconstruction process is carried out on the distance dimension of the radar signal; the clutter suppression/coherent accumulation process is aimed at different pulse echoes of the radar signal. The different pulse echoes are jointly processed; the compression and reconstruction processing is for a single pulse echo of the radar signal, and processes components at different time points in the single echo. 7. a system for realizing claim 1 method, is characterized in that, described system comprises demodulator (1), low-pass filter (2), low-speed ADC (3), burst/parallel converter ( 4), clutter suppression module (5), coherent accumulation module (6) and compression reconstruction module (7); the signal input port of the demodulator (1) is the system signal input port for Receive an analog input signal; the signal output port of the demodulator (1) is connected to the signal input port of the low-pass filter (2); the signal output port of the low-pass filter (2) is connected to the low-speed The signal input port of the ADC (3) is connected; the signal output port of the low-speed ADC (3) is connected with the signal input port of the pulse string/parallel converter (4); the signal of the pulse string/parallel converter (4) The output port is connected to the signal input port (a ₁ ... a _n ) of the clutter suppression module (5); the signal output port of the clutter suppression module (5) is connected to the signal of the coherent accumulation module (6) The input ports (b ₁ ... b _n ) are connected in one-to-one correspondence; the signal output port of the coherent accumulation module (6) is connected with the signal input port (c) of the compression reconstruction module (7); the compression reconstruction module (7) The signal input port (d) of the construction module (7) is used to receive the sparse dictionary signal, and the reconstruction signal output port (e) of the compression reconstruction module (7) is the signal output port of the system; The demodulator (1) is used to mix the pseudo-random sequence and the input analog signal to obtain a mixed frequency signal; the low-pass filter (2) is used to filter out high-frequency interference signals in the mixed frequency signal ; The low-speed ADC (3) is used to sample the input analog signal to obtain a sampled digital signal; the clutter suppression module (5) operates in conjunction with the coherent accumulation module (6), and is used for the clutter suppression of the sampled digital signal /coherent accumulation processing; the compression reconstruction module (7) is used to realize the compression reconstruction of the output signal after the clutter suppression/coherent accumulation processing, and obtain the reconstructed signal. 8. The system according to claim 7, wherein the input analog signal is a radar signal. 9. The system according to claim 8, wherein the compressed sampling step of the radar signal is: Step 1, the radar signal x(t) is mixed with an analog pseudo-random sequence p(t)={1,-1} to obtain the corresponding mixed signal x _p (t)=x(t)·p( t); the bandwidth of the pseudo-random sequence p(t) is the same as the bandwidth of the radar signal x(t); Step 2, using a low-pass filter h(t) to filter the mixed frequency signal to obtain a filtered signal x _h (t)=x(t)*h(t); the low-pass filter h(t) The bandwidth is Fs=500MHz; Step 3, use the low-speed ADC to sample the input analog signal to obtain the sampled digital signal y(m)=x _h (mΔt ₁ ). The sampling frequency of the low-speed ADC is the same as the bandwidth of the low-pass filter h(t), wherein, Δt ₁ =1/Fs is the low-speed ADC sampling interval; Step 4, use the moving target indicator (MTI) and moving target detection (MTD) to perform inter-pulse clutter suppression/coherent accumulation processing to obtain the signal: r′ _il (m)=y _l (m)-2y _l+1 (m)+y _l+2 (m) Wherein, the number of coherent accumulation points of the moving target detection is L=3; the y _l (m) is the compressed sampling result of the lth pulse echo x _l (t) of the radar signal on the slow time dimension ; The r' _il (m) is the result of carrying out second-order cancellation clutter suppression processing to y _l (m); r' _dl (m) is the result of carrying out coherent accumulation processing to y _l (m); W ^nl is a Fourier transform operator; Step 5, using the compression reconstruction algorithm to perform compression reconstruction processing on the obtained signal processed in the step 4 to obtain the reconstruction signal x(n), the reconstruction matrix Φ=H*P*F, where P is p The diagonal matrix that the element in (t) forms, H is the equivalent convolution matrix of analog low-pass filter h (t), and F is the pulse compression matrix of radar; Described radar signal pulse compression point number is I=64. 10. according to the method described in claim 9, it is characterized in that, described radar signal is chirp signal, and radar signal pulse width is B=2GHz; Described radar signal transmission point is N=64, and pulse transmission frequency is PRF= 250MHz, the carrier frequency is Fi=20GHz; the number of targets contained in the radar signal is K=2, and the number of clutter is Kc=4. 11. The method according to claim 9, characterized in that, the moving target indicator (MTI) adopts a two-line delay cancellation mode to eliminate static clutter in the echo signal; the clutter suppression/coherence The accumulation processing is carried out on the pulse dimension of the radar signal; the compression and reconstruction processing is carried out on the distance dimension of the radar signal; the clutter suppression/coherent accumulation processing is aimed at different pulse echoes of the radar signal, and all The different pulse echoes are jointly processed; the compression and reconstruction processing is aimed at a single pulse echo of the radar signal, and processes components at different time points in the single echo. 12. system according to claim 9, is characterized in that, described radar compression sampling system comprises demodulator (1), low-pass filter (2), low-speed ADC (3), burst/parallel converter (4 ), clutter suppression module (5), coherent accumulation module (6) and compression reconstruction module (7); the signal input port of the demodulator (1) is the system signal input port for receiving Analog input signal; the signal output port of the demodulator (1) is connected to the signal input port of the low-pass filter (2); the signal output port of the low-pass filter (2) is connected to the low-speed ADC (3) is connected to the signal input port; the signal output port of the low-speed ADC (3) is connected to the signal input port of the pulse string/parallel converter (4); the signal output of the pulse string/parallel converter (4) The port is connected to the signal input port (a ₁ ... a _n ) of the clutter suppression module (5); the signal output port of the clutter suppression module (5) is connected to the signal input port of the coherent accumulation module (6) The ports (b ₁ ... b _n ) are connected in one-to-one correspondence; the signal output port of the coherent accumulation module (6) is connected to the signal input port (c) of the compression and reconstruction module (7); the compression and reconstruction The signal input port (d) of the module (7) is used to receive the sparse dictionary signal, and the reconstruction signal output port (e) of the compression reconstruction module (7) is the signal output port of the system; The demodulator (1) is used for mixing the pseudo-random sequence and the input radar signal to obtain a mixed frequency signal; the low-pass filter (2) is used for filtering out high-frequency interference signals in the mixed frequency signal ; The low-speed ADC (3) is used to sample the input radar signal to obtain a sampled digital signal; the clutter suppression module (5) operates in conjunction with the coherent accumulation module (6), and is used for the clutter suppression of the sampled digital signal / coherent accumulation processing; the compression reconstruction module (7) is used to realize the compression reconstruction of the output signal after clutter suppression / coherent accumulation processing, and obtain the reconstructed signal; The above modules are running, the specific steps are as follows: Step 1, the radar signal x(t) is mixed with an analog pseudo-random sequence p(t)={1,-1} to obtain the corresponding mixed signal x _p (t)=x(t)·p( t); the bandwidth of the pseudo-random sequence p(t) is the same as the bandwidth of the radar signal x(t); Step 2, using a low-pass filter h(t) to filter the mixed frequency signal to obtain a filtered signal x _h (t)=x(t)*h(t); the low-pass filter h(t) The bandwidth is Fs=500MHz; Step 3, use the low-speed ADC to sample the input analog signal to obtain the sampled digital signal y(m)=x _h (mΔt ₁ ). The sampling frequency of the low-speed ADC is the same as the bandwidth of the low-pass filter h(t), wherein, Δt ₁ =1/Fs is the low-speed ADC sampling interval; Step 4, use the moving target indicator (MTI) and moving target detection (MTD) to perform inter-pulse clutter suppression/coherent accumulation processing to obtain the signal: r′ _il (m)=y _l (m)-2y _l+1 (m)+y _l+2 (m) Wherein, the number of coherent accumulation points of the moving target detection is L=3; the y _l (m) is the compressed sampling result of the lth pulse echo x _l (t) of the radar signal on the slow time dimension ; The r' _il (m) is the result of carrying out second-order cancellation clutter suppression processing to y _l (m); r' _dl (m) is the result of carrying out coherent accumulation processing to y _l (m); W ^nl is a Fourier transform operator; Step 5, using the compression reconstruction algorithm to perform compression reconstruction processing on the obtained signal processed in the step 4 to obtain the reconstruction signal x(n), the reconstruction matrix Φ=H*P*F, where P is p The diagonal matrix that the element in (t) forms, H is the equivalent convolution matrix of analog low-pass filter h (t), and F is the pulse compression matrix of radar; Described radar signal pulse compression point number is I=64.