Method, device and equipment for clutter suppression of strong motion by double SVD reconstruction and storage medium

Abstract

The application discloses a clutter suppression method and device for strong motion, a device, and a storage medium. The method comprises the following steps: constructing a Hankel matrix according to a received echo signal of a radar, performing singular value decomposition and fast Fourier transform to obtain each singular vector spectrum; judging a spectrum correlation coefficient of each singular vector spectrum and a first singular vector spectrum, and judging a spectrum waveform entropy of a singular vector with a high spectrum correlation coefficient value; constructing a Hankel matrix, performing singular value decomposition and fast Fourier transform on a singular vector with a large spectrum waveform entropy value; obtaining a new singular vector through reconstruction of a Hankel matrix and a first inverse Hankel matrix; and reconstructing a Hankel matrix, a second inverse Hankel matrix and a singular vector with a low spectrum correlation coefficient value according to a new diagonal matrix and the new singular vector to obtain a clutter-suppressed echo signal. The application improves the discrimination accuracy of a non-cooperative target under strong motion clutter.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of strong motion clutter suppression, and particularly to a double SVD reconstruction strong motion clutter suppression method, device, equipment and storage medium. BACKGROUND

[0002] With the rapid development of the unmanned aerial vehicle industry, the potential threat of unmanned aerial vehicles is increasing. At the same time, with the continuous growth of flight volume and the continuous improvement of the ecological environment, the airport bird strike prevention work is under increasing pressure. Birds and unmanned aerial vehicles are typical "low, slow and small" targets with low observability. The radar observation environment may be affected by strong motion clutter such as aircraft and vehicles. Therefore, the strong motion clutter suppression method is of great significance to improve the performance of non-cooperative target monitoring and ensure flight safety.

[0003] Existing clutter suppression methods include adaptive moving target indication (AMTI), CLEAN algorithm, wavelet transform, singular value decomposition, and strong motion clutter suppression method based on singular value decomposition. The strong motion clutter suppression method based on singular value decomposition can preserve weak target characteristics and has been widely concerned by domestic and foreign scholars. Zheng Lin et al. proposed a clutter suppression method based on phase coding and singular value decomposition. The method first whitens the clutter through phase decoding, and then separates the clutter according to the difference in autocorrelation between the whitened clutter and the target. Zheng Jun et al. proposed a Wavelet-SVD algorithm. The algorithm first performs two-dimensional wavelet transform on the original radar data, performs singular value decomposition on the transformed wavelet coefficient matrix, removes large singular values, and reconstructs the echo to achieve clutter suppression. M. Garcia-Fernandez et al. proposed a clutter suppression method for synthetic aperture radar (SAR) images. The method performs singular value decomposition on the SAR image, selects the clutter basis by setting a threshold, constructs the clutter subspace and target subspace, and finally projects to the clutter orthogonal complement space to achieve clutter suppression. Huang Fengqing et al. obtained three statistical characteristics of singular value spectrum distribution, singular vector space correlation and average Doppler frequency based on singular value decomposition. The K-means clustering algorithm is used to adaptively determine the singular vector corresponding to the clutter subspace, and the orthogonal subspace projection is used to suppress the clutter component in the echo signal. Wu Linzhu et al. proposed a first-order difference spectrum-SVD reconstruction method based on singular value decomposition theory. The method uses the extreme points of the first-order difference spectrum of singular values to classify and reconstruct the radar echo signal, and realizes effective separation of radar clutter and targets.

[0004] The inventors have found that the existing singular value decomposition-based clutter suppression method has at least the following defects:

[0005] The selection of singular values and the existence of clutter and target information have certain interlacing, so that part of the singular values contain double information of clutter and target, resulting in loss of target information, which will affect the accuracy of target recognition in the observation environment of non-cooperative target detection radar.

[0006] The information disclosed in this Background section is only for the purpose of increasing the understanding of the general background of the application and should not be taken as an acknowledgement or any form of suggestion that this information forms prior art with regard to the application. SUMMARY

[0007] The purpose of the application is to improve the accuracy of target recognition in the observation environment of non-cooperative target detection radar.

[0008] The application provides a double SVD reconstruction strong motion clutter suppression method, comprising the steps of:

[0009] S11, constructing a Hankel matrix according to the echo signals received by the radar in the distance unit where the target is located within the coherent pulse interval, and representing it as ;

[0010] S12, performing singular value decomposition on the Hankel matrix and performing fast Fourier transform to obtain the spectrum of each singular vector;

[0011] S13, first judging the spectral correlation coefficient of each singular vector spectrum and the first singular vector spectrum, comprising:

[0012] judging the spectral waveform entropy of the singular vector with a high spectral correlation coefficient value;

[0013] retaining the singular vector and singular value of the singular vector with a low spectral correlation coefficient value;

[0014] S14, the judgment of the spectral waveform entropy comprises: constructing a Hankel matrix for the singular vector with a large spectral waveform entropy value; assigning a value of zero to the singular value corresponding to the singular vector with a small spectral waveform entropy value to obtain a new diagonal matrix;

[0015] S15, performing singular value decomposition and fast Fourier transform on the constructed singular vector Hankel matrix with a large spectral waveform entropy value, and secondly judging the spectral correlation coefficient of each singular vector spectrum and the first singular vector spectrum;

[0016] The second judgment of the spectral correlation coefficient of each singular vector spectrum and the first singular vector spectrum comprises:

[0017] assigning a value of zero to the singular value corresponding to the singular vector with a high spectral correlation coefficient value;

[0018] The singular vector with low spectral correlation coefficient value is reserved as a singular vector and a singular value;

[0019] S16, reconstructing a Hankel matrix for each singular vector after judging the spectral correlation coefficient for the second time, and obtaining a new singular vector through the first inverse Hankel matrix;

[0020] S17, reconstructing a Hankel matrix according to the new diagonal matrix obtained in step S14, the new singular vector obtained in step S16 and the singular vector with low spectral correlation coefficient value obtained in step S13 for the first time, and obtaining the echo signal after clutter suppression through the second inverse Hankel matrix.

[0021] Preferably, in the embodiment of the present application, when judging the spectral correlation coefficient for the first time, the singular vector with high spectral correlation coefficient value contains a clutter component.

[0022] Preferably, in the embodiment of the present application, when judging the spectral waveform entropy, the singular vector with large spectral waveform entropy value is a singular vector containing a target and a clutter; and the singular vector with small spectral waveform entropy value is a singular vector containing a clutter.

[0023] Preferably, in the embodiment of the present application, the echo signal received by the radar in the distance unit where the target is located within the coherent pulse interval comprises:

[0024] The radar echo signal after ground clutter suppression in the distance unit where the target is located, specifically comprises:

[0025]

[0026] In the formula, is the radar echo signal after ground clutter suppression in the distance unit where the target is located; is a non-cooperative target radar echo signal; is a strong moving clutter; is noise.

[0027] Preferably, in the embodiment of the present application, the singular value decomposition of the Hankel matrix comprises:

[0028] The singular value decomposition of the Hankel matrix is performed through the following formula to obtain a corresponding diagonal matrix, a left singular matrix and a right singular matrix:

[0029]

[0030] In the formula, is the left singular matrix; is the right singular matrix; is the diagonal matrix.

[0031] Preferably, in the embodiment of the present application, the calculation of each singular vector spectrum comprises:

[0032] Each vector of the left singular matrix after singular value decomposition is converted to the frequency domain by fast Fourier transform according to the following formula:

[0033] ,

[0034] In the formula, is each vector spectrum of the left singular matrix; is each vector of the left singular matrix ; when is odd, ; when is even, ; is the length of the received signal in the distance unit.

[0035] Preferably, in the embodiment of the present application, the new diagonal matrix comprises:

[0036] The probability distribution of the frequency component is calculated:

[0037]

[0038] In the formula, is the probability distribution, is an arbitrary spectrum sequence;

[0039] The formula of the spectrum waveform entropy:

[0040]

[0041] In the formula, is the spectrum waveform entropy.

[0042] In another aspect of the present application, a clutter suppression device for reconstructing strong motion by double SVD is also provided, comprising:

[0043] An echo signal Hankel matrix construction unit is configured to construct a Hankel matrix according to the echo signals received by the radar in the distance unit of the target within the coherent pulse interval, and is denoted as ;

[0044] A singular vector spectrum generation unit is configured to perform singular value decomposition on the Hankel matrix and then perform fast Fourier transform to obtain each singular vector spectrum.

[0045] A spectrum correlation coefficient first determination unit is configured to first determine the spectrum correlation coefficient of each singular vector spectrum and the first singular vector spectrum, and comprises:

[0046] The singular vector with high spectrum correlation coefficient value is used to determine the spectrum waveform entropy;

[0047] The singular vector with low spectrum correlation coefficient value is used to reserve the singular vector and singular value;

[0048] The singular vector Hankel matrix construction and new matrix generation unit are used to determine the spectrum waveform entropy, and include the following steps: constructing the Hankel matrix of the singular vector with high spectrum waveform entropy value; and assigning the singular value corresponding to the singular vector with low spectrum waveform entropy value as zero to obtain a new diagonal matrix;

[0049] The spectrum correlation coefficient secondary determination unit is used to perform singular value decomposition and fast Fourier transform on the singular vector Hankel matrix with high spectrum waveform entropy value, and to secondarily determine the spectrum correlation coefficient of each singular vector spectrum and the first singular vector spectrum;

[0050] The secondary determination of the spectrum correlation coefficient of each singular vector spectrum and the first singular vector spectrum includes:

[0051] The singular value corresponding to the singular vector with high spectrum correlation coefficient value is assigned as zero;

[0052] The singular vector with low spectrum correlation coefficient value is used to reserve the singular vector and singular value;

[0053] The new singular vector generation unit is used to reconstruct the Hankel matrix of each singular vector after the secondary determination of the spectrum correlation coefficient, and to obtain a new singular vector through the first inverse Hankel matrix;

[0054] The new echo signal generation unit is used to reconstruct the Hankel matrix of the singular vector with low spectrum correlation coefficient value determined by the first spectrum correlation coefficient determination unit of the singular vector Hankel matrix construction and new matrix generation unit, the new singular vector of the new singular vector generation unit, and the second inverse Hankel matrix to obtain the clutter suppressed echo signal.

[0055] In another aspect of the embodiment of the present application, a double SVD reconstruction strong motion clutter suppression device is also provided. The double SVD reconstruction strong motion clutter suppression device includes a computer program stored on a medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the method described in the above aspects and achieves the same technical effect.

[0056] In another aspect of the embodiment of the present application, a storage medium is also provided, and the storage medium stores a computer program. When the computer program is executed by a processor, each step of the double SVD reconstruction strong motion clutter suppression method according to any one of the above aspects is implemented.

[0057] Compared with the prior art, the present application has the following beneficial effects:

[0058] In the present application, firstly, a Hankel matrix is constructed according to the echo signals received by the radar in the distance unit where the target is located within the coherent pulse interval; after singular value decomposition of the Hankel matrix and fast Fourier transform, the spectrum of each singular vector is obtained; secondly, the spectral correlation coefficient of each singular vector spectrum and the first singular vector spectrum is judged, the spectral waveform entropy of the singular vector with high spectral correlation coefficient value is judged, and the singular vector and singular value of the singular vector with low spectral correlation coefficient value are reserved; then, the Hankel matrix is constructed for the singular vector with large spectral waveform entropy value and singular value decomposition and fast Fourier transform are performed again; the Hankel matrix is reconstructed for each singular vector, and new singular vectors are obtained through the first inverse Hankel matrix; finally, the Hankel matrix is reconstructed according to the new diagonal matrix, the new singular vector and the singular vector with low spectral correlation coefficient value judged first, and the echo signal after clutter suppression is obtained through the second inverse Hankel matrix; the present application utilizes the characteristics of strong power of moving clutter and strong spectral correlation, realizes the suppression of strong moving clutter, avoids the problems of possible residual clutter or loss of part of the target signal compared with the first-order difference spectrum-SVD reconstruction clutter suppression method, can suppress the clutter on the premise of retaining the target micro-motion characteristics, greatly improves the discrimination accuracy of non-cooperative targets in the strong moving clutter background, and can improve the performance of airport bird repelling and unmanned aerial vehicle countermeasure equipment, non-cooperative target monitoring equipment, has commercial value and practical engineering application value for ensuring flight safety.

[0059] The above description is only a summary of the technical solutions of the present application. In order to more clearly understand the technical means of the present application and can be implemented according to the content of the specification, at the same time, in order to make the above and other purposes, technical features and advantages of the present application more easily understood, one or more preferred embodiments are listed below, and the details are described below with the help of the drawings. BRIEF DESCRIPTION OF DRAWINGS

[0060] In order to more clearly illustrate the technical solutions of the present application, the drawings needed for the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creating laborious work.

[0061] Figure 1 is a step diagram of the double SVD reconstruction strong moving clutter suppression method described in the present application;

[0062] Figure 2 is a flow chart of the double SVD reconstruction strong moving clutter suppression method described in the present application;

[0063] Figure 3 is a simulation data unmanned aerial vehicle clutter suppression experimental result graph described in the application;

[0064] Figure 4 is a simulation data bird clutter suppression experimental result graph described in the application;

[0065] Figure 5 is a simulation data two kinds of clutter suppression method before and after the relative power change comparison graph described in the application;

[0066] Figure 6 is a structure schematic diagram of the double SVD reconstruction strong motion clutter suppression device described in the application;

[0067] Figure 7 is a structure schematic diagram of the double SVD reconstruction strong motion clutter suppression device described in the application. DETAILED DESCRIPTION

[0068] The specific embodiments of the application will be described in detail below with reference to the accompanying drawings, but it should be understood that the scope of protection of the application is not limited by the specific embodiments.

[0069] Unless otherwise explicitly stated, throughout the specification and claims, the term "comprise"

[0070] Or its transformation such as "contain" or "include" and the like will be understood to include the stated element or component, and does not exclude other elements or other components.

[0071] In this paper, the terms "first", "second", etc. are used to distinguish two different elements or parts, and are not used to limit the specific position or relative relationship. In other words, in some embodiments, the terms "first", "second", etc. can also be interchangeable with each other.

[0072] Embodiment one

[0073] In order to improve the accuracy of target discrimination in the observation environment of non-cooperative target detection radar, as shown in Figure 1 and Figure 2 In the embodiments of the application, a double SVD reconstruction strong motion clutter suppression method is provided, comprising the steps of:

[0074] S11, according to the echo signal received by the radar in the distance unit where the target is located within the coherent pulse interval, a Hankel matrix is constructed;

[0075] In the embodiments of the application, the non-cooperative target includes "low, slow and small" targets such as flying birds and unmanned aerial vehicles, and strong motion clutter such as airplanes and cars.

[0076] The echo signal received by the radar of the distance unit where the target is located within the coherent pulse interval, comprising:

[0077] The radar echo signal of the distance unit where the target is located after ground clutter suppression, specifically:

[0078]

[0079] In the formula, The radar echo signal of the distance unit where the target is located after ground clutter suppression; The non-cooperative target radar echo signal; The strong motion clutter; The noise.

[0080] Due to the similar spectrum shape and center frequency of the strong motion clutter components, the clutter components have the characteristics of spectrum correlation, so as to utilize the spectrum correlation of the strong motion clutter, first convert the Hankel matrix of the clutter components to a two-dimensional Hankel matrix; let , the Hankle matrix of which is can be written as:

[0081]

[0082] In the formula, The length of the signal received by the resolution unit to be processed, ; when is odd, ; when is even, ; determined by the equation ; Each row in the matrix has elements same as the previous row, and the correlation between rows is strong.

[0083] S12, singular value decomposition is performed on the Hankel matrix, and fast Fourier transform is performed, to obtain the spectrum of each singular vector;

[0084] In the embodiment of the application, singular value decomposition is performed on the Hankel matrix by the following formula, to obtain the corresponding diagonal matrix, left singular matrix and right singular matrix:

[0085]

[0086] In the formula, The left singular matrix; The right singular matrix; The diagonal matrix.

[0087] The left singular matrix , the right singular matrix is an orthogonal matrix, is a diagonal matrix composed of corresponding singular values after singular value decomposition:

[0088] ,

[0089] is the number of singular values of is a zero matrix, and the diagonal matrix The values on the diagonal are sorted in descending order. is a diagonal matrix the first singular value of After singular value decomposition, the singular value and singular vector contain all the information of the echo signal.

[0090] In the embodiment of the application, the calculation of the spectrum of each singular vector includes:

[0091] Each vector of the left singular matrix after singular value decomposition is converted to the frequency domain by fast Fourier transform according to the following formula:

[0092] ,

[0093] In the formula, is the spectrum of each vector of the left singular matrix; is each vector of the left singular matrix When is odd, When is even, ; is the length of the received signal in the distance unit.

[0094] In the embodiment of the application, considering that the clutter power is much larger than the target power, the left singular vector corresponding to the first singular value is regarded as a clutter component; the spectrum of the left and right singular vectors is symmetric about zero frequency, so only the spectrum of the left singular vector and the correlation of the first left singular vector spectrum are analyzed.

[0095] S13, first determine the spectral correlation coefficient of each singular vector spectrum and the first singular vector spectrum, including:

[0096] For singular vectors with high spectral correlation coefficient values, determine the spectral waveform entropy;

[0097] For singular vectors with low spectral correlation coefficient values, the singular vector and singular value are retained;

[0098] In the embodiment of the present application, when the spectral correlation coefficient is first determined, the singular vector with a high spectral correlation coefficient value contains clutter components; the singular vector with a large correlation coefficient value is extracted by using the spectral correlation coefficient (Pearsen correlation coefficient) for quantitative analysis of the correlation degree, and the singular vector mainly contains clutter components.

[0099] S14, the spectral waveform entropy is determined, including: constructing a Hankel matrix for the singular vector with a large spectral waveform entropy value; assigning a singular value corresponding to the singular vector with a small spectral waveform entropy value as zero to obtain a new diagonal matrix;

[0100] In the embodiment of the present application, when the spectral waveform entropy is determined, the singular vector with a large spectral waveform entropy value is a singular vector containing a target and clutter; and the singular vector with a small spectral waveform entropy value is a singular vector containing clutter.

[0101] Specifically, the spectral waveform entropy can describe the complexity and randomness of the signal spectrum distribution, and the larger the value is, the more frequency components contained in the signal; therefore, the singular vector mainly containing clutter components obtained in step S13 is divided into a singular vector containing only clutter and a singular vector containing a target and clutter by using the spectral waveform entropy, the singular vector with a large spectral waveform entropy value is considered as a singular vector containing a target and clutter, and the singular vector with a small spectral waveform entropy value is considered as a singular vector containing only clutter. The singular value corresponding to the singular vector containing only clutter is assigned as zero to obtain a new diagonal matrix.

[0102] The new diagonal matrix includes:

[0103] The probability distribution of the frequency component is calculated:

[0104]

[0105] In the formula, is the probability distribution, is an arbitrary spectral sequence;

[0106] The formula of the spectral waveform entropy is:

[0107]

[0108] In the formula, is the spectral waveform entropy.

[0109] S15, singular value decomposition and fast Fourier transform are performed on the constructed singular vector Hankel matrix with a large spectral waveform entropy value, and the spectral correlation coefficient of each singular vector spectrum and the first singular vector spectrum is determined again;

[0110] The spectral correlation coefficient of each singular vector spectrum and the first singular vector spectrum is determined again, including:

[0111] The singular value corresponding to the singular vector with high spectrum correlation coefficient value is assigned as zero;

[0112] The singular vector and singular value of the singular vector with low spectrum correlation coefficient value are reserved;

[0113] In the embodiment of the application, in order to reserve more complete target micro-motion components, the Hankel matrix of the singular vector containing the target and the clutter is constructed in the step S14, singular value decomposition (SVD) is performed, the Pearson correlation coefficient of the spectrum of all singular vectors and the spectrum of the first singular vector (the singular vector corresponding to the maximum singular value) is calculated, the singular vector with high Pearson correlation coefficient value is selected, and the singular value corresponding to the singular vector is assigned as zero.

[0114] S16, reconstruct the Hankel matrix of each singular vector after the secondary judgment spectrum correlation coefficient, and obtain new singular vectors through the first inverse Hankel matrix;

[0115] In the embodiment of the application, the Hankel matrix is reconstructed, the new singular vector is obtained by using the inverse Hankel transformation, and the new singular vector replaces the singular vector containing the target and the clutter to obtain new left and right singular matrices And , the specific process is as follows:

[0116] Unlike the first SVD reconstruction, the Hankel matrix is constructed for the left and right singular vectors containing the target and the clutter, that is,

[0117]

[0118] In the formula, (taking the left singular vector as an example), the SVD decomposition is performed on to obtain , and The Fourier transformation is performed on the left singular vector, the Pearson correlation coefficient of the spectrum of the left singular vector and the spectrum of the first left singular vector is calculated, the singular value corresponding to the singular vector with a larger correlation coefficient value is assigned as zero, and a new diagonal matrix is obtained, the Hankel matrix is restored by using the singular value with a smaller correlation coefficient and the corresponding singular vector, that is,

[0119]

[0120] The singular vector containing only the target micro-motion information is restored by the inverse Hankel transformation, the left and right singular vectors containing only the target micro-motion information are used to replace the original left and right singular vectors containing the clutter and the target micro-motion information, and new left and right singular matrices are obtained and .

[0121] S17. Reconstruct the Hankel matrix based on the new diagonal matrix, the new singular vector, and the singular vector with the lowest initial spectral correlation coefficient, and obtain the echo signal after clutter suppression through the second inverse Hankel matrix.

[0122] In this embodiment of the invention, the diagonal matrix obtained in step S14 is used and the left and right singular matrices obtained in step S16 and The Hankel matrix is ​​reconstructed, and the clutter-suppressed target signal is obtained through inverse Hankel transform; the specific process is as follows:

[0123] Using a new diagonal matrix Left and right singular matrices and Reconstruct a new Hankel matrix ,Right now

[0124]

[0125] In the formula, To reconstruct the elements in the Hankel matrix, an inverse Hankel transformation is required. The inverse Hankel transformation process is shown in the following formula:

[0126]

[0127] Signal after clutter suppression for:

[0128]

[0129] In this embodiment of the invention, radar parameters, bird parameters, and UAV parameters are shown in Tables 1, 2, and 3.

[0130] Table 1:

[0131]

[0132] Table 2:

[0133]

[0134] Table 3:

[0135]

[0136] The clutter suppression effect of the dual SVD reconstruction of the present invention is as follows: Figure 3 , Figure 4 and Figure 5 As shown:

[0137] Figure 3 Figure 1 is a chart of experimental results of UAV clutter suppression, wherein Figure 3 a is a UAV received signal spectrum, Figure 3 b is a UAV spectrum after clutter suppression (first-order difference spectrum-SVD reconstruction), Figure 3 c is a UAV spectrum after clutter suppression;

[0138] Figure 4 Figure 2 is a chart of experimental results of bird clutter suppression, wherein Figure 4 a is a bird received signal spectrum, Figure 4 b is a bird spectrum after clutter suppression (first-order difference spectrum-SVD reconstruction), Figure 4 c is a bird spectrum after clutter suppression; relative to a target received signal spectrum chart, from Figure 3 b, 3c, Figure 4 b, 4c, it can be seen that after the first-order difference spectrum-SVD reconstruction method is used for clutter suppression, there is a problem of residual clutter, and the present application can suppress the moving clutter while retaining relatively complete target information.

[0139] In order to quantitatively evaluate the clutter suppression performance, a relative change amount of power after clutter suppression and target echo power is defined, that is,

[0140]

[0141] In the formula, is a bird or UAV echo amplitude without considering clutter, is a bird or UAV echo amplitude after clutter suppression, and the smaller the change amount is, the better the clutter suppression effect is; the relative power change after clutter suppression by the first-order difference spectrum-SVD reconstruction method and the present application is as shown in Figure 5 With the increase of signal-to-clutter ratio, the relative power change amount is first reduced to a certain extent and then tends to be flat, and compared with the first-order difference spectrum-SVD reconstruction method, the relative power change amount of the present application is smaller.

[0142] To sum up, according to the embodiment of the present application, firstly, a Hankel matrix is constructed according to echo signals received by the radar in the distance unit where the target is located within the coherent pulse interval; after singular value decomposition of the Hankel matrix and fast Fourier transform, the spectrum of each singular vector is obtained; secondly, the spectral correlation coefficient of each singular vector spectrum and the first singular vector spectrum is judged, the spectral waveform entropy of the singular vector with a high spectral correlation coefficient value is judged, and the singular vector and singular value of the singular vector with a low spectral correlation coefficient value are reserved; then, the singular vector with a large spectral waveform entropy value is used to construct a Hankel matrix and perform singular value decomposition and fast Fourier transform again; the Hankel matrix is reconstructed according to each singular vector, and a new singular vector is obtained through the first inverse Hankel matrix; finally, the Hankel matrix is reconstructed according to the new diagonal matrix, the new singular vector and the singular vector with a low spectral correlation coefficient value judged in the first time, and the echo signal after clutter suppression is obtained through the second inverse Hankel matrix; the present application utilizes the characteristics of strong motion clutter power and strong spectral correlation to realize the suppression of strong motion clutter, avoids the problems of possible residual clutter or loss of part of the target signal compared with the first-order difference spectrum-SVD reconstruction clutter suppression method, can suppress the clutter on the premise of retaining the target micro-motion characteristics, greatly improves the discrimination accuracy of non-cooperative targets in the strong motion clutter background, and can improve the performance of airport bird repelling and unmanned aerial vehicle countermeasure equipment, non-cooperative target monitoring equipment, has commercial value and practical engineering application value for ensuring flight safety.

[0143] Embodiment two

[0144] Corresponding to the method embodiment, in another aspect of the embodiment of the present application, a double SVD reconstruction strong motion clutter suppression device is also provided, Figure 6 The structure schematic diagram of the double SVD reconstruction strong motion clutter suppression device provided by the embodiment of the present application is shown, and the double SVD reconstruction strong motion clutter suppression device is a device corresponding to the double SVD reconstruction strong motion clutter suppression method in the embodiment of the present application, Figure 1 The device corresponding to the double SVD reconstruction strong motion clutter suppression method in the embodiment is realized by a virtual device, that is, Figure 1 The double SVD reconstruction strong motion clutter suppression method in the corresponding embodiment, each virtual module of the double SVD reconstruction strong motion clutter suppression device can be executed by an electronic device, such as a network device, a terminal device or a server. Specifically, the double SVD reconstruction strong motion clutter suppression device in the embodiment of the present application comprises:

[0145] The echo signal Hankel matrix construction unit 01 is used to construct a Hankel matrix according to echo signals received by the radar in the distance unit where the target is located within the coherent pulse interval;

[0146] The echo signal of the radar received by the distance unit where the target is located in the coherent pulse interval comprises:

[0147] The radar echo signal of the distance unit where the target is located after ground clutter suppression is specifically:

[0148]

[0149] In the formula, is the radar echo signal of the distance unit where the target is located after ground clutter suppression; is the radar echo signal of the non-cooperative target of birds or unmanned aerial vehicles; is strong motion clutter; is noise.

[0150] Since the frequency spectrum shape and the center frequency of the strong motion clutter component are similar, the clutter components have the characteristics of frequency spectrum correlation, so as to utilize the frequency spectrum correlation of the strong motion clutter, first, is converted into a two-dimensional Hankel matrix; let , the Hankel matrix of which is can be written as:

[0151]

[0152] In the formula, is the length of the signal received by the resolution unit to be processed, ; when is odd, ; when is even, ; is determined by the equation ; Each row in the equation has elements same as the previous row, and the correlation between rows is strong.

[0153] The singular vector spectrum generation unit 02 is configured to perform singular value decomposition on the Hankel matrix and perform fast Fourier transform to obtain each singular vector spectrum.

[0154] In the embodiment of the application, the singular value decomposition is performed on the Hankel matrix by the following formula to obtain the corresponding diagonal matrix, left singular matrix and right singular matrix:

[0155]

[0156] In the formula, is the left singular matrix; is the right singular matrix; is the diagonal matrix.

[0157] The left singular matrix , right singular matrix is an orthogonal matrix, is a diagonal matrix composed of corresponding singular values after singular value decomposition:

[0158] ,

[0159] is the number of singular values of is a zero matrix, diagonal matrix the values on the diagonal are sorted in descending order; is a diagonal matrix the first singular value of

[0160] In the embodiments of the application, the calculation of each singular vector spectrum includes:

[0161] Each vector of the left singular matrix after singular value decomposition is converted to the frequency domain by fast Fourier transform according to the following formula:

[0162] ,

[0163] In the formula, is the spectrum of each vector of the left singular matrix; is each vector of the left singular matrix ; when is odd, ; when is even, ; is the length of the received signal in the distance unit.

[0164] The spectrum correlation coefficient first determination unit 03 is used to first determine the spectrum correlation coefficient of each singular vector spectrum and the first singular vector spectrum, and includes:

[0165] The spectrum waveform entropy of the singular vector with a high spectrum correlation coefficient value is determined;

[0166] The singular vector with a low spectrum correlation coefficient value is reserved;

[0167] In the embodiments of the application, when the spectrum correlation coefficient is first determined, the singular vector with a high spectrum correlation coefficient value contains a clutter component; the degree of correlation is quantitatively analyzed by using the spectrum correlation coefficient (Pearsen correlation coefficient), and the singular vector with a large correlation coefficient value is extracted, which mainly contains a clutter component.

[0168] The singular vector Hankel matrix construction and new matrix generating unit 04 is configured to judge the spectral waveform entropy, and construct a Hankel matrix for the singular vector with a large spectral waveform entropy value; and assign a value of zero to the singular value corresponding to the singular vector with a small spectral waveform entropy value, so as to obtain a new diagonal matrix.

[0169] In the embodiment of the present application, when judging the spectral waveform entropy, the singular vector with a large spectral waveform entropy value is a singular vector containing a target and clutter; and the singular vector with a small spectral waveform entropy value is a singular vector containing clutter.

[0170] The new diagonal matrix comprises:

[0171] The probability distribution of the frequency component is calculated.

[0172]

[0173] In the formula, P (f) is the probability distribution, is an arbitrary spectrum sequence.

[0174] The formula of the spectral waveform entropy is:

[0175]

[0176] In the formula, P (f) is the probability distribution,

[0177] The spectrum correlation coefficient secondary judgment unit 05 is configured to perform singular value decomposition and fast Fourier transform on the singular vector Hankel matrix with a large spectral waveform entropy value constructed, and secondarily judge the spectrum correlation coefficient of each singular vector spectrum and the first singular vector spectrum.

[0178] The secondarily judging the spectrum correlation coefficient of each singular vector spectrum and the first singular vector spectrum comprises:

[0179] The singular value corresponding to the singular vector with a high spectrum correlation coefficient value is assigned a value of zero.

[0180] The singular vector and the singular value of the singular vector with a low spectrum correlation coefficient value are reserved.

[0181] In the embodiment of the present application, in order to reserve a more complete target micro-motion component, the singular vector Hankel matrix containing the target and the clutter in the constructing step S14 is decomposed by singular value decomposition (SVD), the Pearson correlation coefficient of all singular vector spectrums and the first singular vector (the singular vector corresponding to the maximum singular value) spectrum is calculated, and the singular vector with a high Pearson correlation coefficient value is selected, and the singular value corresponding to the singular vector is assigned a value of zero.

[0182] ​​The new singular vector generating unit 06 is configured to reconstruct a Hankel matrix by using each singular vector after the secondary judgment of the spectrum correlation coefficient, and obtain a new singular vector by the first inverse Hankel matrix.

[0183] In the embodiment of the present application, the Hankel matrix is reconstructed, the new singular vector is obtained by using the inverse Hankel transform, and the new left singular matrix and the right singular matrix are obtained by replacing the singular vector containing the target and the clutter at the same time.

[0184] The new echo signal generating unit 07 is configured to reconstruct a Hankel matrix by using the new diagonal matrix, the new singular vector and the singular vector with a low spectrum correlation coefficient value in the first judgment, and obtain the echo signal after the clutter suppression by the second inverse Hankel matrix.

[0185] In the embodiment of the present application, the diagonal matrix obtained in the step S14 and the left singular matrix and the right singular matrix obtained in the step S16 are used to reconstruct the Hankel matrix, and the target signal after the clutter suppression is obtained by the inverse Hankel transform.

[0186] The new diagonal matrix , the left singular matrix and the right singular matrix and are used to reconstruct the new Hankel matrix , that is

[0187]

[0188] In the formula, is an element in the reconstructed Hankel matrix, and the inverse Hankel transform needs to be performed on

[0189]

[0190] The signal after the clutter suppression is as follows:

[0191]

[0192] It should be noted that the specific implementation manner and technical effects of the double SVD reconstruction strong motion clutter suppression device in the embodiment of the present application can refer to the double SVD reconstruction strong motion clutter suppression method corresponding to Figure 1 , and details are not repeated here.

[0193] Embodiment three

[0194] ​Corresponding to the method embodiment, the embodiment of the application also provides a double SVD reconstruction strong motion clutter suppression device, such as a terminal, a server, etc. The server can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDNs, and basic cloud computing services such as big data and artificial intelligence platforms. The terminal can be a smart phone, a tablet computer, a notebook computer, a desktop computer, etc., but is not limited thereto.

[0195] The hardware structure block diagram of the double SVD reconstruction strong motion clutter suppression device provided by the embodiment of the application is shown in the example diagram as Figure 7

[0196] The processor 1, the communication interface 2, the memory 3, and the communication bus 4;

[0197] The processor 1, the communication interface 2, the memory 3, and the communication bus 4;

[0198] Optionally, the communication interface 2 can be an interface of a communication module, such as an interface of a GSM module.

[0199] The processor 1 can be a central processing unit CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the application.

[0200] The memory 3 can include a high-speed RAM memory, and can also include a non-volatile memory, such as at least one disk memory.

[0201] The processor 1 is specifically configured to execute a computer program stored in the memory 3 to perform the following steps:

[0202] S11, constructing a Hankel matrix according to echo signals received by a radar within a coherent pulse interval and a distance unit where a target is located;

[0203] S12, performing singular value decomposition on the Hankel matrix and performing a fast Fourier transform to obtain a spectrum of each singular vector;

[0204] S13, first judging a spectrum correlation coefficient of each singular vector spectrum and a first singular vector spectrum, including:

[0205] Judging a spectrum waveform entropy of a singular vector with a high spectrum correlation coefficient value.​

[0206] The singular vector and the singular value of the singular vector with a low spectral correlation coefficient value are reserved;

[0207] S14, the spectral waveform entropy of the singular vector with a large spectral waveform entropy value is constructed; the singular value corresponding to the singular vector with a small spectral waveform entropy value is assigned as zero to obtain a new diagonal matrix;

[0208] S15, singular value decomposition and fast Fourier transform are performed on the singular vector Hankel matrix with a large spectral waveform entropy value, and the spectral correlation coefficient of each singular vector spectrum and the first singular vector spectrum is judged again;

[0209] The spectral correlation coefficient of each singular vector spectrum and the first singular vector spectrum is judged again, including:

[0210] The singular value corresponding to the singular vector with a high spectral correlation coefficient value is assigned as zero;

[0211] The singular vector and the singular value of the singular vector with a low spectral correlation coefficient value are reserved;

[0212] S16, the Hankel matrix of each singular vector after the spectral correlation coefficient is judged again is reconstructed, and a new singular vector is obtained through the first inverse Hankel matrix;

[0213] S17, the Hankel matrix of the new diagonal matrix, the new singular vector and the singular vector with a low spectral correlation coefficient value is reconstructed, and the echo signal after clutter suppression is obtained through the second inverse Hankel matrix.

[0214] The above product can execute the method provided by the embodiment of the application, has the corresponding function module and beneficial effects of executing the method. Technical details not described in detail in the embodiment can be referred to the double SVD reconstruction strong motion clutter suppression method provided by the embodiment of the application.

[0215] Embodiment four

[0216] In the embodiment of the application, a storage medium is also provided, which can store a program suitable for processor execution, and the program is used for:

[0217] S11, a Hankel matrix is constructed according to the echo signal received by the radar in the distance unit where the target is located within the coherent pulse interval;

[0218] S12, singular value decomposition is performed on the Hankel matrix, and fast Fourier transform is performed to obtain the spectrum of each singular vector;

[0219] S13, judging the spectrum correlation coefficient of each singular vector spectrum and the first singular vector spectrum, comprising:

[0220] judging the spectrum waveform entropy of the singular vector with high spectrum correlation coefficient value;

[0221] retaining the singular vector and singular value of the singular vector with low spectrum correlation coefficient value;

[0222] S14, judging the spectrum waveform entropy, constructing a Hankel matrix for the singular vector with large spectrum waveform entropy value, assigning zero to the singular value corresponding to the singular vector with small spectrum waveform entropy value, and obtaining a new diagonal matrix;

[0223] S15, performing singular value decomposition and fast Fourier transform on the constructed Hankel matrix of the singular vector with large spectrum waveform entropy value, and twice judging the spectrum correlation coefficient of each singular vector spectrum and the first singular vector spectrum;

[0224] the twice judging the spectrum correlation coefficient of each singular vector spectrum and the first singular vector spectrum, comprising:

[0225] assigning zero to the singular value corresponding to the singular vector with high spectrum correlation coefficient value;

[0226] retaining the singular vector and singular value of the singular vector with low spectrum correlation coefficient value;

[0227] S16, reconstructing a Hankel matrix for each singular vector after twice judging the spectrum correlation coefficient, and obtaining a new singular vector through the first inverse Hankel matrix;

[0228] S17, reconstructing a Hankel matrix according to the new diagonal matrix, the new singular vector, and the singular vector with low spectrum correlation coefficient value in the first judgment, and obtaining the echo signal after clutter suppression through the second inverse Hankel matrix.

[0229] Optionally, the refinement function and the expansion function of the program can refer to the description above.

[0230] The product described above can execute the method provided by the embodiments of the application, and has the corresponding function modules and beneficial effects of the execution method. Technical details not described in detail in the embodiments can refer to the method provided by other embodiments of the application.

[0231] Those skilled in the art can understand that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0232] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. In addition, the coupling or direct coupling or communication connection between the displayed or discussed units can be indirect coupling or communication connection through some interface, device or unit, and can be electrical, mechanical or other forms.

[0233] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, i.e. they can be located in one place or distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0234] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically, or two or more units can be integrated into one unit.

[0235] It should be understood that the features in the embodiments of the present application, each embodiment, feature can be combined with each other, and can achieve the purpose of solving the above technical problems.

[0236] If the functions are realized in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the parts of the technical solutions that essentially contribute to the prior art can be embodied in the form of software products, which are stored in a storage medium and include a number of instructions for making a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present application. The foregoing storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk and various program code storage media.

[0237] The foregoing description of the disclosed embodiments enables a person skilled in the art to make or use the application. Modifications of these embodiments will occur to persons of skill in the art, and that the generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Therefore, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A clutter suppression method for strong motion reconstructed by dual SVD, characterized in that, Including the following steps: S11. Construct the Hankel matrix based on the echo signal received by the radar of the target's range cell within the coherent pulse interval, denoted as: ; S12. After performing singular value decomposition on the Hankel matrix and then performing a fast Fourier transform, the spectrum of each singular vector is obtained. S13. The first determination of the spectral correlation coefficient between each singular vector spectrum and the first singular vector spectrum includes: For singular vectors with high spectral correlation coefficients, determine the spectral waveform entropy; For singular vectors with low spectral correlation coefficients, retain both the singular vector and the singular value; S14. The determination of the spectrum waveform entropy includes: constructing a Hankel matrix for the singular vectors with large spectrum waveform entropy values; assigning zero to the singular values ​​corresponding to the singular vectors with small spectrum waveform entropy values ​​to obtain a new diagonal matrix; S15. Perform singular value decomposition and fast Fourier transform on the singular vector Hankel matrix with large entropy value of the constructed spectrum waveform, and determine the spectral correlation coefficient between each singular vector spectrum and the first singular vector spectrum for the second time. The secondary determination of the spectral correlation coefficient between each singular vector spectrum and the first singular vector spectrum includes: The singular values ​​corresponding to singular vectors with high spectral correlation coefficients are assigned the value of zero. For singular vectors with low spectral correlation coefficients, both the singular vector and the singular value are retained. S16. Reconstruct the Hankel matrix for each singular vector after the second judgment of the spectral correlation coefficient, and obtain the new singular vector through the first inverse Hankel matrix; S17. Reconstruct the Hankel matrix based on the new diagonal matrix obtained in step S14, the new singular vector obtained in step S16, and the singular vector with low initial spectral correlation coefficient obtained in step S13, and obtain the echo signal after clutter suppression through the second inverse Hankel matrix.

2. The clutter suppression method for strong motion reconstructed by dual SVD according to claim 1, characterized in that, When initially determining the spectral correlation coefficient, the singular vector with a high spectral correlation coefficient value contains clutter components.

3. The clutter suppression method for strong motion reconstructed by dual SVD according to claim 2, characterized in that, When determining the spectral waveform entropy, the singular vector with a large spectral waveform entropy value is a singular vector containing both the target and clutter; the singular vector with a small spectral waveform entropy value is a singular vector containing clutter.

4. The clutter suppression method for strong motion reconstructed by dual SVD according to claim 1, characterized in that, The echo signal received by the radar of the range cell where the target is located within the coherent pulse interval includes: The radar echo signal of the target's range cell after ground clutter suppression is as follows: ； In the formula, This is the radar echo signal of the target's range cell after ground clutter suppression; This is a radar echo signal from a non-cooperative target; Strong motion clutter; It is noise.

5. The clutter suppression method for strong motion reconstructed by dual SVD according to claim 4, characterized in that, The singular value decomposition of the Hankel matrix includes: The Hankel matrix is ​​decomposed into singular values ​​using the following formulas to obtain the corresponding diagonal matrix, left singular matrix, and right singular matrix: ； In the formula, It is a left singular matrix; It is a right singular matrix; It is a diagonal matrix.

6. The clutter suppression method for strong motion reconstructed by dual SVD according to claim 5, characterized in that, The calculation of the spectrum of each singular vector includes: The vectors of the left singular matrix after singular value decomposition are transformed to the frequency domain by fast Fourier transform using the following formula; ， ； In the formula, The vector spectra of the left singular matrix; Left singular matrix Each vector; when It is an odd number. ;when When it is even, ; The length of the signal received by the unit at that distance.

7. The clutter suppression method for strong motion reconstructed by dual SVD according to claim 1, characterized in that, The new diagonal matrix includes: Calculate the probability distribution of the frequency components: ； In the formula, For probability distribution, It is an arbitrary spectral sequence; The formula for the entropy of the spectral waveform is: ； In the formula, It represents the entropy of the spectral waveform.

8. A clutter suppression device for strong motion reconstructed by dual SVD, characterized in that, include: The echo signal Hankel matrix construction unit is used to construct a Hankel matrix based on the echo signal received by the radar of the target's range cell within the coherent pulse interval, denoted as: ; The singular vector spectrum generation unit is used to perform singular value decomposition on the Hankel matrix and then perform a fast Fourier transform to obtain the spectrum of each singular vector. The initial determination unit for spectral correlation coefficient is used to initially determine the spectral correlation coefficient between each singular vector spectrum and the first singular vector spectrum, including: For singular vectors with high spectral correlation coefficients, determine the spectral waveform entropy; For singular vectors with low spectral correlation coefficients, retain both the singular vector and the singular value; The singular vector Hankel matrix construction and new matrix generation unit is used to determine the spectral waveform entropy, including: constructing a Hankel matrix for singular vectors with large spectral waveform entropy values; assigning zero to the singular values ​​corresponding to singular vectors with small spectral waveform entropy values ​​to obtain a new diagonal matrix; The secondary determination unit for spectral correlation coefficient is used to perform singular value decomposition and fast Fourier transform on the constructed singular vector Hankel matrix with large spectral waveform entropy value, and to determine the spectral correlation coefficient between each singular vector spectrum and the first singular vector spectrum. The secondary determination of the spectral correlation coefficient between each singular vector spectrum and the first singular vector spectrum includes: The singular values ​​corresponding to singular vectors with high spectral correlation coefficients are assigned the value of zero. For singular vectors with low spectral correlation coefficients, both the singular vector and the singular value are retained. The new singular vector generation unit is used to reconstruct the Hankel matrix of each singular vector after the second judgment of the spectral correlation coefficient, and obtain the new singular vector through the first inverse Hankel matrix; The new echo signal generation unit is used to reconstruct the Hankel matrix based on the singular vector Hankel matrix, the new diagonal matrix of the new matrix generation unit, the new singular vector of the new singular vector generation unit, and the singular vector with low spectral correlation coefficient value in the first determination unit of spectral correlation coefficient, and to obtain the clutter-suppressed echo signal through the second inverse Hankel matrix.

9. A clutter suppression device for strong motion reconstructed by dual SVD, characterized in that, include: Memory, used to store computer programs; A processor is configured to invoke and execute the computer program to implement the steps of the dual SVD reconstruction clutter suppression method for strong motion as described in any one of claims 1-7.

10. A storage medium, characterized in that, Includes a software program adapted by a processor to perform the steps of the dual SVD reconstruction of strong motion clutter suppression method as described in any one of claims 1-7.

Images