Frequency spectrum dispersion interference suppression method based on time-frequency image processing

Abstract

The invention provides a frequency spectrum dispersion interference suppression method based on time-frequency image processing, which comprises the following steps of: firstly, carrying out short-time Fourier transform on two pulse receiving signals with different frequency offsets sent by using a frequency control array radar to generate a two-dimensional time-frequency image containing time domain and frequency domain information; the method comprises the following steps of: firstly, acquiring two time-frequency images, then eliminating the same part, namely interference components, of the two time-frequency images through pixel difference operation, retaining the difference part, namely target components, ensuring the difference by carrier frequency change in a frequency-phase cooperative mode, and finally, restoring a fast time signal through inverse short-time Fourier transform, thereby laying a foundation for the extraction of subsequent target information. According to the scheme, the frequency spectrum dispersion interference and the target echo are effectively distinguished and suppressed by using the carrier frequency difference between the pulses of the frequency-phase cooperative frequency control array radar, priori estimation on the frequency spectrum dispersion interference is not needed, interference identification and elimination can be completed only through two pulses, meanwhile, target components can be effectively reserved, and high practicability is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of radar active interference suppression, and in particular to a spectrum dispersion interference suppression method based on time-frequency image processing. Background Art

[0002] Smeared Spectrum (SMSP) jamming is a jamming technique for linear frequency modulation (LFM) radars. It down-converts radar pulses to a clock frequency f. s / N performs analog-to-digital conversion, then stores the data, copies it, and re-reads it at the clock frequency f s Perform digital-to-analog conversion, then up-convert and transmit to generate a sub-pulse with a frequency modulation slope N times that of the original signal and a pulse width 1 / N of the original intercepted pulse, and replicate the sub-pulse N times to generate an interference signal with the same pulse width as the original signal. Spectral dispersion interference can form a large number of dense false targets before and after the real target, for the purpose of deception and concealment in the distance unit.

[0003] At present, the countermeasures against spectral dispersion interference are mainly divided into two categories: interference signal separation and interference signal cancellation after reconstruction. In the existing signal separation algorithm, it is usually necessary to reconstruct the target echo in combination with compressed sensing theory after separation. However, in the case of low signal-to-noise ratio and high interference-to-signal ratio, it is easy to cause problems such as large interference residue and mismatch of signal reconstruction dictionary. In the research on cancellation of interference signal after reconstruction, it is necessary to estimate the characteristic parameters of different interference signals. The accuracy of frequency modulation coefficient estimation of spectral dispersion interference has a great influence on the interference suppression effect. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention uses a frequency-controlled array (FDA) radar based on frequency deviation coordination to suppress spectrum dispersion interference using a time-frequency image generated by short-time Fourier transform (STFT), and proposes a spectrum dispersion interference suppression method based on time-frequency image processing. The method comprises:

[0005] Step S1: Use a frequency-controlled array radar to send pulse signals with different frequency offsets, and perform STFT processing and analysis on the received echoes;

[0006] Step S2: according to the processing result of STFT, the spectrum dispersion interference is eliminated by using the image processing method, and the obtained image data matrix is ​​subjected to inverse short-time Fourier transform to restore the target component;

[0007] Step S3: The restored time domain signal is processed by a multi-channel mixer-matched filter structure receiver, and the receiver uses a Multiple Signal Classification (MUSIC) algorithm to perform spatial spectrum estimation to determine the true signal position.

[0008] Furthermore, the step S1 includes the following steps:

[0009] Step S11: The frequency-controlled array radar transmits a first pulse signal with a frequency deviation of Δf1. The jammer mounted on the target intercepts the pulse signal and performs spectrum dispersion processing on it. Then, the jammer signal and the target echo are sent together to a radar receiver that uses N array elements to receive signals. The obtained echo signal is subjected to STFT analysis to obtain a time-frequency image data matrix Y1 and a corresponding power spectrum P1.

[0010] Step S12: the frequency-controlled array radar transmits a second pulse signal with a frequency deviation of Δf2, and the receiver performs STFT analysis on the received signal including the target echo and the interference signal to obtain a time-frequency image data matrix Y2 and its corresponding power spectrum P2;

[0011] Step S13: Compare the time-frequency images of the echo pulses of the two pulses. If the same information exists in the time-frequency images of the two pulses, it means that there is an interference component, and the part of the echo information that changes with the frequency deviation, that is, the difference part, is the real target. Because for the real target, the time-frequency images of the pulses corresponding to different frequency deviations are obviously different.

[0012] Furthermore, the step S2 comprises the following steps:

[0013] Step S21: adjusting the elimination coefficient k according to the change of the interference intensity between the two pulses, then subtracting the time-frequency image Y1 corresponding to the first pulse signal adjusted with the elimination coefficient from the time-frequency image Y2 corresponding to the second pulse signal to obtain a difference time-frequency image, and subtracting the power spectrum P1 corresponding to the first pulse signal adjusted with the elimination coefficient from the power spectrum P2 corresponding to the second pulse signal to obtain a difference power spectrum;

[0014] Step S22: setting the data with difference power spectrum value less than 0 to 0, and the time-frequency data matrix of the corresponding position to 0;

[0015] Step S23: performing an inverse STFT transformation on the processed difference time-frequency data matrix to obtain a time domain signal after interference suppression;

[0016] Further, the S3 comprises the following steps:

[0017] Step S31: performing eigendecomposition on the covariance matrix of the interference suppressed time domain signal through a multi-channel mixer-matched filter structure receiver;

[0018] Step S32: Obtain the number of signal sources in the signal according to Akaike Information Criterion (AIC);

[0019] Step S33: Use the MUSIC algorithm to perform spatial spectrum search to obtain estimated values ​​of the azimuth and distance of the radar detection target. There is no interference information in the result, and the interference suppression is successful.

[0020] The beneficial technical effect of the present invention is as follows: the present invention transmits two detection signals with different frequency deviations through a frequency-controlled array radar, utilizes the characteristics that the information of the forwarding interference such as spectrum dispersion interference is related to the frequency deviation information of a certain echo copied by it, and uses STFT to analyze and process the echo information of the two detection signals to obtain the time-frequency image and power spectrum of the two echo signals, and then uses the image processing method to eliminate the spectrum dispersion interference, that is, by subtracting the time-frequency image and power spectrum corresponding to the first pulse signal from the time-frequency image and power spectrum corresponding to the second pulse signal, thereby removing the interference component therein, and then restores the time domain signal through the inverse STFT transformation of the data after the interference elimination processing, and then uses the multi-channel mixing-matched filter for processing, and finally verifies that there is no interference component in the signal through the MUSIC algorithm. The scheme of the present invention utilizes the carrier frequency difference between pulses of the frequency-phase coordinated frequency-controlled array radar, effectively realizes the distinction and suppression of spectrum dispersion interference and target echo, does not need to make a priori estimation of spectrum dispersion interference, and only needs two pulses to complete the identification and elimination of interference, and can effectively retain the target component at the same time, improve the efficiency and accuracy of interference suppression, and has high practicality. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0022] Figure 1 It is a flow chart of a method for suppressing spectral dispersion interference based on time-frequency image processing provided by an embodiment of the present invention;

[0023] Figure 2 It is a frequency domain image of an interference signal with a transmission frequency deviation of Δf1 and a target echo provided by an embodiment of the present invention;

[0024] Figure 3It is a frequency domain image of an interference signal with a transmission frequency deviation of Δf2 and a target echo provided by an embodiment of the present invention;

[0025] Figure 4 It is a time-frequency analysis diagram of an interference signal and a target echo with a transmission frequency deviation of Δf1 provided by an embodiment of the present invention;

[0026] Figure 5 It is a time-frequency analysis diagram of an interference signal and a target echo with a transmission frequency deviation of Δf2 provided by an embodiment of the present invention;

[0027] Figure 6 is a time-frequency image of a difference value after image processing provided by an embodiment of the present invention;

[0028] Figure 7 It is a distance estimation based on the MUSIC algorithm of the signal after inverse STFT transformation provided by an embodiment of the present invention;

[0029] Figure 8 This is the angle estimation of the MUSIC algorithm based on the signal after inverse STFT transformation provided by the embodiment of the present invention. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] like Figure 1 As shown, the present invention proposes a method for suppressing spectral dispersion interference based on time-frequency image processing, comprising the following steps:

[0032] Step S1: Use a frequency-controlled array radar with M array elements and d intervals to transmit a linear frequency modulation signal. K is the frequency modulation coefficient, T p is the pulse width. For the first pulse, the carrier frequencies of the signals transmitted by each element of the frequency-controlled array are f 1,m =f c +mΔf1, where f c is the reference carrier frequency, m=0,1,...,M-1, represents the array element number, Δf1 is the carrier frequency deviation, then the first pulse signal transmitted by the mth array element is:

[0033]

[0034] Consider a far-field target located at (r0, θ0), where r0 is the radial distance of the target and θ0 is the azimuth of the target. The self-defense jammer mounted on it intercepts the pulse signal and performs spectrum dispersion processing to generate a spectrum dispersion interference signal x spmp (t), the expression is:

[0035]

[0036] Where N represents the frequency modulation coefficient of the spectrum dispersion interference. Considering the carrier, the signal actually forwarded by the jammer is:

[0037]

[0038] Then the first pulse echo signal received by the radar receiver is expressed as:

[0039]

[0040] in, represents the round-trip time delay of the echo, c is the speed of light, n(t) is the propagation noise, and the received echo signal y1(t) is subjected to short-time Fourier transform of discrete signals. The corresponding STFT is expressed as:

[0041]

[0042] Among them, j and They correspond to the number of sampling points of the signal in the time dimension and frequency dimension respectively, w(·) is the window function of STFT processing, L is the moving step size of the window function, N t is the total length of the discrete-time signal, and t = t1, t2, ... t Nt , we can see that the signal to be analyzed y1(t) is cut into multiple segments by the window function, and Fourier transform is performed on each segment. Taking the square of the modulus gives the energy density or power spectrum of y1(t):

[0043]

[0044] The results of STFT are affected by the window function and the transmitted waveform. Since the main research here is the impact of frequency deviation of the frequency-controlled array on the interference signal, the common Hamming window is selected.

[0045] Step S2: The radio-controlled array transmits a second pulse signal with a frequency deviation of Δf2. At this time, the carrier frequency of each array element is f 2,m =f c+mΔf2, since the enemy jammer cannot predict the variable carrier frequency strategy of our radar and the spectrum dispersion interference modulation is relatively complex, the carrier frequency information of the spectrum dispersion interference in the echo signal at this stage is the same as the carrier frequency information of the first pulse it intercepted. The echo signal of the pulse signal is expressed as:

[0046]

[0047] Among them, ξ s and j Respectively represent the loss constants related to the amplitude of the target echo and the interference signal, and in this embodiment, their values ​​are 1;

[0048] The obtained echo signal y2(t) is subjected to short-time Fourier transform processing of discrete signals, and its corresponding STFT is expressed as:

[0049]

[0050] Likewise, Taking the square of the modulus gives the energy density or power spectrum of y2(t):

[0051]

[0052] Step S3: Compare the time-frequency diagrams of the echo pulses of the two pulses. If the same information can be clearly observed in the time-frequency diagrams of the two pulses, there is an interference component. The part of the echo information that changes with the frequency deviation is the real target, because for the real target, the time-frequency diagrams of the pulses corresponding to different frequency deviations are obviously different.

[0053] Step S4: The time-frequency image (data matrix) corresponding to the second pulse signal (frequency deviation is Δf2) is subtracted from the time-frequency image corresponding to the first pulse (frequency deviation is Δf1) adjusted by the elimination coefficient to obtain a difference time-frequency image. The power spectrum corresponding to the second pulse signal (frequency deviation is Δf2) is subtracted from the power spectrum corresponding to the first pulse (frequency deviation is Δf1) adjusted by the elimination coefficient to obtain a difference power spectrum. The difference time-frequency image and the difference power spectrum are respectively expressed as:

[0054]

[0055] Among them, k is the elimination coefficient, and the elimination coefficient is adjusted according to the change of the interference intensity between two pulses. Under normal circumstances, the interference intensity between pulses does not change much and is usually taken as 1.

[0056] Step S5: Set the data with difference power spectrum value less than 0 to 0, and also set the time-frequency data matrix of the corresponding position to 0.

[0057] Step S6: Process the difference time-frequency data matrix Perform inverse STFT transform to obtain time domain data y diff (t).

[0058] Step S7: Set y diff (t) is processed by a multi-channel mixer-matched filter structure receiver. For the unit to be detected τ′, we have:

[0059]

[0060] in,(·) * represents the conjugate operation, ae,2(t) represents the array time-domain steering vector of the second pulse, expressed as:

[0061]

[0062] Since the receiving end uses N array elements to receive signals, the echo signal received by the detection unit τ′ is expressed as:

[0063]

[0064] In the formula

[0065]

[0066] r j is the radial distance of the disturbance, θ j is the azimuth of interference, n represents noise, R s (Δτ′) is the autocorrelation function of the transmitted pulse signal, a R,x (θ0) represents the receiving steering vector, a T,x (r,θ) represents the transmission steering vector of the array, x represents the pulse number, which is 1 or 2 in this embodiment, and Δτ′=τ0-τ′ represents the difference between the target delay and the sampling delay of the unit to be detected. Usually, the signal and the noise signal are independent of each other, and the signal covariance matrix is ​​eigen-decomposed, and then:

[0067] R ydiff,2 =Σ2Λ2Σ2=Σ s2 Λ s2 Σ s2 +Σ n2 Λ n2 Σ n2

[0068] In the formula, Σ s2 ,Σ n2 are the signal subspace and noise subspace of the echo signal with a transmission frequency deviation of Δf2, Λ s2 and Λ n2 is the positive definite matrix corresponding to the signal and the noise.

[0069] Step S8: Using the signal subspace, the number of signal sources is obtained according to the Akaike Information Criterion (AIC), providing a basis for subsequent Multiple Signal Classification (MUSIC) estimation.

[0070] Step S9: Use the MUSIC algorithm to perform a spatial spectrum search with a frequency deviation of Δf2 to obtain an estimated value of the azimuth and distance of the radar detection target, specifically:

[0071]

[0072] in,(·) H represents taking the conjugate transpose, It is the transmit-receive joint steering vector of the frequency-controlled array radar. There is no interference information in the current result, and the interference suppression is successful.

[0073] The spectral dispersion interference suppression method based on time-frequency image processing is experimentally verified, and the experimental parameters are set as follows:

[0074] Table 1 Parameter settings

[0075] parameter value Carrier frequency 1.005GHz <![CDATA[Frequency offset Δf1]]> 10MHz <![CDATA[Frequency offset Δf2]]> 5MHz Pulse Width 50us bandwidth 5MHz Sampling rate 200MHz Interference-to-noise ratio 30dB Signal-to-Noise Ratio 30dB Pulse repetition frequency 200us Target Information (10°,12km) Speed ​​​​deception jamming information (10°,12km)

[0076] The experimental results are as follows Figure 2-Figure 8 As shown, Figure 2 It is the frequency domain image of the interference signal with a transmission frequency deviation of Δf1 and the target echo. It can be seen from the figure that the frequency characteristics of the target signal and the spectrum dispersion interference are consistent, while in terms of intensity, the target signal is submerged in the spectrum dispersion interference.

[0077] Figure 3 It is the frequency domain image of the interference signal with a transmission frequency deviation of Δf2 and the target echo. It can be seen from the figure that the frequency characteristics of the target signal and the spectrum dispersion interference are different, but the target signal intensity is still weaker than the spectrum dispersion interference, making it difficult to extract the target information.

[0078] Figure 4 It is a time-frequency analysis diagram of the interference signal with a transmission frequency deviation of Δf1 and the target echo. The image mainly shows the characteristics of spectrum dispersion interference, and the target signal is submerged in it.

[0079] Figure 5 It is a time-frequency analysis diagram of the interference signal with a transmit frequency deviation of Δf2 and the target echo. The image simultaneously presents the time-frequency characteristics of the spectrum dispersion interference and the target signal. However, due to the change of the carrier frequency, the target component is obviously different from the spectrum dispersion interference.

[0080] Figure 6It is a time-frequency image of the difference after image processing. It can be seen from the figure that the target component in the image is relatively intact, while the component of spectral diffusion interference has been greatly weakened.

[0081] Figure 7 It is the distance estimation of the MUSIC algorithm based on the signal after inverse STFT transformation. It can be seen from the figure that the image only reflects the distance information of the target and no longer has the component of spectrum dispersion interference.

[0082] Figure 8 It is the angle estimation based on the MUSIC algorithm of the signal after inverse STFT transformation. It can be seen from the figure that the image only reflects the angle information of the target and no longer has the component of spectrum dispersion interference.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A spectrum dispersion interference suppression method based on time-frequency image processing, characterized in that: The method comprises: Step S1: Use a frequency-controlled array radar to send pulse signals with different frequency offsets, and perform short-time Fourier transform (STFT) processing and analysis on the received echoes; Step S2: according to the processing result of STFT, the spectrum dispersion interference is eliminated by using the image processing method, and the obtained image data matrix is ​​subjected to inverse short-time Fourier transform to restore the target component; Step S3: The restored time domain signal is processed by a multi-channel mixer-matched filter structure receiver, and a multiple signal classification (MUSIC) algorithm is used in the receiver to perform spatial spectrum estimation to determine the true signal position.

2. The method according to claim 1, characterized in that The step S1 further comprises: Step S11: The frequency-controlled array radar transmits a first pulse signal with a frequency deviation of Δf1, and the jammer mounted on the target intercepts the pulse signal and performs spectrum dispersion processing on it. Then, the jammer signal and the target echo are sent together to a radar receiver that uses N array elements to receive signals, and the obtained echo signal is subjected to STFT processing to obtain a time-frequency image Y1 and a corresponding power spectrum P1; Step S12: the frequency-controlled array radar transmits a second pulse signal with a frequency deviation of Δf2, and the receiver performs STFT analysis on the received signal including the target echo and the interference signal to obtain a time-frequency image Y2 and a corresponding power spectrum P2; Step S13: Compare the time-frequency images of the echo pulses of the two pulses. If the same information exists in the time-frequency images of the two pulses, it means that there is an interference component, and the part of the echo information that changes with the frequency deviation, that is, the difference part, is the real target.

3. The method according to claim 1, characterized in that The step S2 further comprises: Step S21: subtract the time-frequency image Y1 corresponding to the first pulse signal adjusted by the elimination coefficient from the time-frequency image Y2 corresponding to the second pulse signal to obtain a difference time-frequency image, and subtract the power spectrum P1 corresponding to the first pulse signal adjusted by the elimination coefficient from the power spectrum P2 corresponding to the second pulse signal to obtain a difference power spectrum; Step S22: setting the data with difference power spectrum value less than 0 to 0, and the time-frequency data matrix of the corresponding position to 0; Step S23: performing an inverse STFT transform on the processed difference time-frequency data matrix to obtain a time domain signal after interference suppression.

4. The method according to claim 1, characterized in that: The step S3 further comprises: Step S31: performing eigendecomposition on the covariance matrix of the interference suppressed time domain signal through a multi-channel mixer-matched filter structure receiver; Step S32: obtaining the number of signal sources in the signal according to the Akaike Information Criterion AIC; Step S33: Use the MUSIC algorithm to perform spatial spectrum search to obtain estimated values ​​of the azimuth and distance of the radar detection target. There is no interference information in the result, and the interference suppression is successful.

5. The method according to claim 2, characterized in that: In step S11, the first pulse echo signal received by the radar receiver is expressed as: Among them, the signal K is the frequency modulation coefficient, T p is the pulse width. For the first pulse, the carrier frequencies of the signals transmitted by each element of the frequency-controlled array are f 1,m =f c +mΔf1,f c is the reference carrier frequency, m=0,1,...,M-1, represents the array element number, Δf1 is the carrier frequency deviation, x spmp (t) is the spectrum dispersion interference signal, N is the frequency modulation coefficient of the spectrum dispersion interference, represents the round-trip time delay of the echo, c is the speed of light, and n(t) is the propagation noise.

6. The method according to claim 5, characterized in that The echo signal is processed by STFT to obtain the time-frequency image, which is expressed as: Among them, j and l F They correspond to the number of sampling points of the signal in the time dimension and frequency dimension respectively, w(·) is the window function of STFT processing, L is the moving step size of the window function, N t is the total length of the signal in discrete time, and For Y1(j,l F ) takes the square of the modulus to obtain the energy density or power spectrum of y1(t): P1(j,l F )=|Y1(j,l F )| 2 .

7. The method according to claim 6, characterized in that In step S12, the echo signal received after sending the second pulse signal is expressed as: Among them, ξ s and j Represent the loss constants related to the amplitude of the target echo and interference signal respectively.

8. The method according to claim 7, characterized in that The echo signal y2(t) is processed by STFT and expressed as: For Y2(j,l F ) takes the square of the modulus to obtain the energy density of y2(t), i.e., the power spectrum: P2(j,l F )=|Y2(j,l F )| 2 .

9. The method according to claim 8, characterized in that The difference time-frequency image and difference power spectrum obtained in step S21 are respectively expressed as: Y diff (j,l F )=Y2(j,l F )-k·Y1(j,l F ) P diff (j,l F )=P2(j,l F )-k·P2(j,l F ) The elimination coefficient k is adjusted according to the change of the interference strength between two pulses.

Images