A radar distributed target detection method, system, device and medium

Abstract

The present invention discloses a radar distributed target detection method, system, device and medium, and relates to the field of radar target detection technology. The method includes: obtaining first radar data of distributed targets in a to-be-detected area, performing data whitening operation on the first radar data to obtain second radar data; determining an HPD matrix according to the second radar data, and determining the eigenvalue of the HPD matrix according to the HPD matrix; determining a first distributed target detector through a correspondence between the eigenvalue and the distributed target detector, obtaining a detection result according to the first distributed target detector, and determining whether there is a distributed target according to a comparison result between the detection result and a preset value. The present invention can expand the detection of targets, greatly improve the accuracy of the detection results, and first perform a whitening operation on the data received by the radar to suppress clutter, thereby improving the detection probability. Moreover, under different signal energy distribution conditions, the present invention also has a certain robustness.

Description

Background Art

[0002] Radar target detection plays a vital role in maritime surveillance, but sea clutter often exhibits non-Gaussian and non-uniform characteristics during target detection, especially in high-resolution radar systems.

[0003] Multifunctional and multi-task radars usually use shorter pulse sequences to improve processing speed. However, traditional constant false alarm rate (CFAR) detectors, such as cell-averaging CFAR (CA-CFAR) using fast Fourier transform (FFT), may cause Doppler resolution degradation and energy leakage when processing data from short pulse sequences. Information geometry transforms the target detection problem into the problem of distinguishing targets from clutter on a manifold, which does not require prior information about clutter and has advantages when the number of pulses is small. However, most matrix CFAR detectors based on information geometry currently have relatively large computational complexity, and the detection probability of the detector is not very ideal. Moreover, in the actual environment, in addition to noise and possible signals, there are also a lot of interference, which may be caused by intentional electronic countermeasures or unintentional process production, which further exacerbates the difficulty of radar target detection in the case of short pulse sequences. Summary of the invention

[0004] The technical problem to be solved by the present invention is to address the deficiencies of the prior art and specifically provide a radar distributed target detection method, system, device and medium, which are as follows:

[0005] 1) In the first aspect, the present invention provides a radar distributed target detection, and the specific technical solution is as follows:

[0006] S1, acquiring first radar data of distributed targets in a to-be-detected area, and performing a data whitening operation on the first radar data to obtain second radar data;

[0007] S2, determining a HPD matrix according to the second radar data, and determining an eigenvalue of the HPD matrix according to the HPD matrix;

[0008] S3, determining the first distributed target detector corresponding to the feature value through the correspondence between the feature value and the distributed target detector, obtaining a detection result according to the first distributed target detector, and determining whether a distributed target exists according to a comparison result between the detection result and a preset value.

[0009] The beneficial effects of the radar distributed target detection provided by the present invention are as follows:

[0010] In the case of short pulses and the target and clutter spectrum overlap, the method of the present invention can be applied to the detection of extended targets, greatly improving the accuracy of the detection results. The data received by the radar is first whitened to suppress clutter, thereby improving the detection probability. Compared with existing detectors, its computational complexity is lower than that of general information geometry matrix CFAR detectors, and under different signal energy distribution conditions, the present invention also has a certain robustness.

[0011] Based on the above solution, the present invention can also be improved as follows.

[0012] Further, the first radar data includes:

[0013] At least two of the target signal, clutter and Gaussian white noise.

[0014] Furthermore, the data whitening operation is specifically as follows:

[0015]

[0016]

[0017] in, represents the sample covariance matrix, K represents the number of units under test, L represents the number of training samples collected around the units under test, when l = 1, ..., K, z l Indicates the measured data, represents the whitened measured data. When l=K+1,…,K+L, z l represents the training sample data, represents the whitened training samples, Indicates z l Perform a conjugate transpose operation.

[0018] Furthermore, the corresponding relationship between the eigenvalue and the distributed target detector is:

[0019]

[0020] Among them, t a,DT is the statistics of the detector, represents the HPD matrix, Indicates taking Maximum value, represents the eigenvalue of the HPD matrix, a represents the power parameter, N is the number of pulses in the CPI, and n=1,...,N represents the sequence number of the eigenvalue.

[0021] 2) In the second aspect, the present invention also provides a radar distributed target detection system, and the specific technical solution is as follows:

[0022] An acquisition module, used for acquiring first radar data of distributed targets in a to-be-detected area, and performing a data whitening operation on the first radar data to obtain second radar data;

[0023] a calculation module, configured to determine a HPD matrix according to the second radar data, and determine an eigenvalue of the HPD matrix according to the HPD matrix;

[0024] The detector module is used to determine the first distributed target detector corresponding to the feature value through the correspondence between the feature value and the distributed target detector, obtain the detection result according to the first distributed target detector, and determine whether there is a distributed target according to the comparison result between the detection result and the preset value.

[0025] Based on the above solution, the present invention can also be improved as follows.

[0026] Further, the first radar data includes:

[0027] At least two of the target signal, clutter and Gaussian white noise.

[0028] Furthermore, the data whitening operation is specifically as follows:

[0029]

[0030]

[0031] in, represents the sample covariance matrix, K represents the number of units under test, L represents the number of training samples collected around the units under test, when l=1,...,K, z l Indicates the measured data, represents the whitened measured data. When l=K+1,...,K+L, z l represents the training sample data, represents the whitened training samples, Indicates z l Perform a conjugate transpose operation.

[0032] Furthermore, the corresponding relationship between the eigenvalue and the distributed target detector is:

[0033]

[0034] Among them, t a,DT is the statistics of the detector, represents the HPD matrix, Indicates taking Maximum value, represents the eigenvalue of the HPD matrix, a represents the power parameter, N is the number of pulses in the CPI, and n=1,...,N represents the sequence number of the eigenvalue.

[0035] 3) In a third aspect, the present invention further provides a computer device, comprising a processor, wherein the processor is coupled to a memory, wherein the memory stores at least one computer program, and wherein the at least one computer program is loaded and executed by the processor so that the computer device implements any of the above methods.

[0036] 4) In a fourth aspect, the present invention further provides a computer-readable storage medium, wherein at least one computer program is stored in the computer-readable storage medium, and the at least one computer program is loaded and executed by a processor so that a computer implements any of the above methods.

[0037] It should be noted that the beneficial effects achieved by the technical solutions of the second to fourth aspects of the present invention and the corresponding possible implementation methods can be found in the above-mentioned technical effects of the first aspect and its corresponding possible implementation methods, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0039] Figure 1 A schematic diagram of a process flow of an embodiment of the present invention;

[0040] Figure 2 It is a schematic diagram of the variation of PD with output SCR of the whitened detector under different signal models 1 to 4 according to an embodiment of the present invention;

[0041] Figure 3 It is a schematic diagram of signal mode 1 of the variation of PD of the whitened detector and the maximum eigenvalue detector with the output SCR according to an embodiment of the present invention;

[0042] Figure 4 It is a schematic diagram of signal mode 2 of the variation of PD of the whitened detector and the maximum eigenvalue detector with the output SCR according to an embodiment of the present invention;

[0043] Figure 5 It is a schematic diagram of signal mode 3 of the variation of PD of the whitened detector and the maximum eigenvalue detector with the output SCR according to an embodiment of the present invention;

[0044] Figure 6 It is a schematic diagram of signal mode 4 of the variation of PD of the whitened detector and the maximum eigenvalue detector with the output SCR according to an embodiment of the present invention;

[0045] Figure 7 It is a schematic diagram of the covariance matrix of the PD of the whitened detector and the non-whitened detector according to the embodiment of the present invention with a lag-one correlation coefficient of 0.9;

[0046] Figure 8 It is a schematic diagram of the covariance matrix of the PD of the whitened detector and the non-whitened detector according to the embodiment of the present invention with a lag-one correlation coefficient of 0.75;

[0047] Fig. 9 It is a schematic diagram of the variation of the clutter output SCR according to the lag-one correlation coefficient ρ of the covariance matrix according to an embodiment of the present invention;

[0048] Fig.10 It is a schematic diagram of the variation of PD with output SCR of a whitened detector and a non-whitened detector in an embodiment of the present invention using real sea clutter data;

[0049] Fig.11 Schematic diagram of a device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0051] like Figure 1 As shown, a radar distributed target detection method according to an embodiment of the present invention comprises the following steps:

[0052] S1, acquiring first radar data of distributed targets in a to-be-detected area, and performing a data whitening operation on the first radar data to obtain second radar data;

[0053] S2, determining a HPD matrix according to the second radar data, and determining an eigenvalue of the HPD matrix according to the HPD matrix;

[0054] S3, determining the first distributed target detector corresponding to the feature value through the correspondence between the feature value and the distributed target detector, obtaining a detection result according to the first distributed target detector, and determining whether a distributed target exists according to a comparison result between the detection result and a preset value.

[0055] The beneficial effects of a radar distributed target detection method provided by the present invention are as follows:

[0056] In the case of short pulses and the target and clutter spectrum overlap, the method of the present invention can be applied to the detection of extended targets, greatly improving the accuracy of the detection results. The data received by the radar is first whitened to suppress clutter, thereby improving the detection probability. Compared with existing detectors, its computational complexity is lower than that of general information geometry matrix CFAR detectors, and under different signal energy distribution conditions, the present invention also has a certain robustness.

[0057] S1, obtaining first radar data of distributed targets in a to-be-detected area, and performing a data whitening operation on the first radar data to obtain second radar data.

[0058] The area to be detected may be: a sea surface where targets may exist (the targets may be distributed targets or point targets).

[0059] Distributed targets are: in the area to be detected, the targets that occupy multiple distance units in the range resolution of the radar are called distributed targets, that is, the number of cells under test (CUT) (positive integer) K>1; if K=1, the detected target is modeled as a point target. Among them, the range resolution refers to the ability of the radar system to distinguish two or more targets at the same direction but different distances. The range resolution of the radar can be changed. The range resolution of the radar is affected by many factors, such as pulse width, receiver sensitivity, target size and shape, propagation characteristics of electromagnetic waves, etc. The degree of range resolution mainly depends on the width of the transmitted pulse. The range unit refers to the small grid that the radar range resolution divides the space into in the distance latitude. A small grid is called a range unit. The measured unit refers to the distance resolution unit occupied by the target to be detected, which is equivalent to the concept of the distance unit.

[0060] In this embodiment, the calculation is mainly based on distributed targets, which can solve the detection of larger targets and improve accuracy. It can also be used for the detection of point targets. Distributed targets can be fleets, etc., and point targets can be small fishing boats, etc.

[0061] In this embodiment, the target (i.e., the main data or the measured data) located in the measured unit is represented by a vector z l ,l=1,...,K means that since the target occupies K distance units in the radar range resolution, the target is a distributed target; if the target occupies 1 distance unit in the radar range resolution, the target is a point target. In this case, K=1 and the target is represented by vector z l (l=1). At the same time, L training samples (i.e. auxiliary data or training sample data) are collected on both sides of the CUT, and the vector z l ,l=K+1,...,K+L represents.

[0062] The process from radar generating echo to forming received data is as follows:

[0063] The radar will first emit a specific electromagnetic wave signal, which will propagate in the air. When encountering a target, part of the electromagnetic wave will be reflected back. The radar antenna will receive the reflected electromagnetic wave, which is the so-called echo. The received echo signal will be converted into an electrical signal and then sent to the input of the receiver. The received electrical signal will undergo a series of processing, such as amplification, filtering, detection, etc., to eliminate noise and improve the signal-to-noise ratio, so as to obtain a clear echo signal. The processed echo signal will be converted into a digital signal and then sent to the computer for further processing and analysis. In the computer, the received digital signal will be further analyzed and processed, such as distance measurement, speed measurement, angle measurement, etc., and finally various detailed information about the target will be formed.

[0064] In coherent processing, the signal accumulation and enhancement is usually carried out by using the interval pulse number method. The interval pulse number refers to the number of pulses at which the processing is performed during coherent processing.

[0065] The first radar data can be: after the radar transmits a radar pulse to the area to be detected, the radar echo is received, and the radar data received by the receiver is further processed and analyzed (such as the above-mentioned received electrical signal will undergo a series of processing, such as amplification, filtering, detection, etc., to eliminate noise and improve the signal-to-noise ratio, so as to obtain a clear echo signal. The processed echo signal will be converted into a digital signal) to obtain the first radar data (digital signal). Simply put, the radar echo is the original, unprocessed signal, and the data received by the radar is the information after processing and analysis. When the number of pulses in the coherent processing interval (CPI) is N, the radar echo of the lth distance unit can be represented by the N-dimensional complex vector z l =[z 0,l ,...,z N-1,l ] T ,l=1,...,K+L means, [z 0,l ,...,z N-1,l ] T Indicates z 0,l ,...,z N-1,l Perform a vector transpose operation. In simple terms, it can be understood as z l =[z 0,l ,...,z N-1,l ] T ,l=1,...,K+L is expressed as (when l=1,...,K, z l Represents the measured data. When l=K+1,...,K+L, zl represents the summary data of the training sample data), that is, the first radar data. In this embodiment, for the occurrence of z l The specific meaning is distinguished by the value of the subscript l, such as: when l=1,...,K, z l Represents the measured data. When l=K+1,...,K+L, z l represents the training sample data. When l=1,...,K,...,K+L, z l Represents the summary of the measured data and the training sample data, that is, the first radar data.

[0066] Therefore, the detection of targets can be divided into: with target and without target. The first formula can be used to describe the target detection situation. The first formula is as follows:

[0067]

[0068] Among them, z l represents the first radar data received for the lth range unit, and the target signal is represented by s l =β l p,l=1,...,K, where β l is the unknown target amplitude, represents the steering vector of the signal; f d is the normalized target Doppler frequency; c l represents clutter, n l Represents Gaussian white noise; H0 represents the hypothesis when there is no target, and H1 represents the hypothesis when there is a target.

[0069] Clutter l It obeys the composite Gaussian distribution and is determined by the second formula, which is as follows:

[0070]

[0071] Among them, τ l represents the slowly varying texture component, η l represents the fast-changing speckle component, the speckle component η l is an independent, zero-mean, complex circular Gaussian random vector, whose normalized covariance matrix R is N×N dimensional (because the number of CPI pulses is N); when a coherent integration interval is short, the texture component τ of each distance unit l It can be approximately regarded as a random constant, in which case the composite Gaussian distribution can be transformed into a spherically invariant random process (SIRP).

[0072] The first radar data includes at least two of a target signal, clutter, and Gaussian white noise.

[0073] When there is a target in the area to be detected, the first radar data includes: target signal, clutter and Gaussian white noise; when there is no target in the area to be detected, the first radar data includes: clutter and Gaussian white noise.

[0074] The first radar data is subjected to data whitening through a sample covariance matrix. The sample covariance matrix is ​​specifically determined by a second formula, which is as follows:

[0075]

[0076] in, represents the sample covariance matrix. When l=1,...,K, z l Represents the measured data. When l=K+1,...,K+L, z l represents the training sample data, Indicates z l Perform a conjugate transpose operation.

[0077] The second radar data is determined by the third formula, which is as follows:

[0078]

[0079] in, represents the sample covariance matrix. When l=1,...,K, z l Indicates the measured data, represents the whitened measured data. When l=K+1,…,K+L, z l represents the training sample data, Represents the whitened training sample data.

[0080] The beneficial effects of adopting the above scheme are as follows:

[0081] The data received by the radar is whitened to suppress clutter and thus increase the probability of detection.

[0082] S2, determining an HPD matrix according to the second radar data, and determining an eigenvalue of the HPD matrix according to the HPD matrix.

[0083] The HPD matrix is ​​determined by the fourth formula, which is as follows:

[0084]

[0085] in, represents the HPD matrix, Representation Statistics Expectations; (·) * represents conjugation, such as Indicates r 1，l Perform conjugation operation; The expression representing the second radar data is used to convert Substituting here for calculation should not be understood as any other new meaning. The specific calculation process is as follows: Formula 5 and Formula 6; N is the number of pulses in CPI, r m,l ,0≤m≤N-1 is the correlation coefficient, which is determined by the fifth formula, which is as follows:

[0086]

[0087] To solve the matrix According to the ergodicity of the stationary Gaussian process, the correlation coefficient r is estimated by averaging the whitened data m,l , the average whitening data is determined by the sixth formula, which is as follows:

[0088]

[0089] in, is the estimated value of the correlation coefficient, yes The qth element of So the value range of q here is 0≤q≤N).

[0090] according to Substituting the value of into the fourth formula, the corresponding HPD matrix is ​​obtained According to the HPD matrix Get the eigenvalues ​​of the HPD matrix Specifically eig represents the HPD matrix Take the eigenvalue.

[0091] S3, determining a first distributed target detector corresponding to the feature value through the correspondence between the feature value and the distributed target detector, obtaining a detection result according to the first distributed target detector, and determining whether there is a distributed target according to a comparison result between the detection result and a preset value.

[0092] The corresponding relationship between the eigenvalue and the distributed target detector is determined by the seventh formula, which is as follows:

[0093]

[0094] in, Indicates taking Maximum value, Represents the eigenvalue of the matrix, according to Calculated; n = 1, ..., N represents the sequence number of the eigenvalue, HPD matrix It is N×N dimensional and has N eigenvalues; a represents a power parameter; by adjusting different power parameters a, different first distributed target detectors can be obtained; it is specifically determined by the eighth formula, which is as follows:

[0095]

[0096] in, Indicates taking Minimum value; t a,DT is the statistic of the detector; when a=-1,1,0,-∞,∞, the corresponding detectors can be called DT-MHM-DW, DT-MAM-DW, DT-MGM-DW, DT-MME-DW and DT-MEMD-DW respectively, and the corresponding abbreviations are explained as follows: Distributed Target (DT); Data Whitening (DW); Maximum Eigenvalue to Harmonic mean (MHM); Maximum Eigenvalue to Arithmetic mean (MAM); Maximum Eigenvalue to Geometric mean (MGM); Maximum Eigenvalue to Minimum eigenvalue (MME); Maximum Eigenvalue-based Matrix Detection (MEMD).

[0097] In the eighth formula, the detector with a = -1 has the best detection performance, but the detection probability is similar to that of the detector with a = 0, 1, ∞, +∞. However, in the design process of the detector, a data whitening process is used, which makes the detection performance of the detector using the data whitening process proposed by the present invention better and obvious compared with the detector without data whitening.

[0098] When the detector statistic t a,DT When it is greater than the threshold determined by the false alarm probability, it is considered that there is a target, otherwise, there is no target.

[0099] The false alarm probability refers to the probability that the detector will falsely sound an alarm when there is no target. For example, under the assumption of H0 (i.e., when there is no target), the detector indicates that there is a target. This is called a false alarm. In 1000 simulation experiments, there was one case where the detector indicated that there was a target when there was no target. In the remaining 999 experiments, the detector correctly indicated that there was no target, indicating that the false alarm probability is 10. -3 .

[0100] The following is an experiment performed on the method proposed in this embodiment:

[0101] The Monte Carlo method is used to evaluate the performance of the detector proposed in Formula 8. In order to shorten the duration of the experiment, the false alarm probability PFA is set to 10 -3 100 / PFA independent experiments were used to determine the threshold, and 5000 independent experiments were used to obtain PD.

[0102] SCR in :The input SCR usually refers to the ratio of the target signal to the background clutter. In a radar system, the target signal usually comes from the radar wave reflected by the target object, while the background clutter may come from a variety of noise sources, such as weather changes, equipment failures, etc. The larger the input SCR, the stronger the target signal is and the greater the possibility of the target object's existence.

[0103] SCR out : The output SCR refers to the ratio of the target signal output by the detector to the background clutter. After the detector processes the input signal, the ratio of the output signal strength to the background clutter is the output SCR. The larger the output SCR, the greater the possibility that the detector successfully detects the target object.

[0104] Given a set of signal-to-noise ratios SCR, calculate the target amplitude β, substitute β into the first formula, and perform subsequent calculations to obtain the experimental results.

[0105] The input and output SCR are expressed as:

[0106]

[0107]

[0108] Among them, tr(R) represents the trace of the covariance matrix R, and the covariance matrix R has the following structure:

[0109] R=R0+p0I N ,

[0110] in, Among them, exp(·) represents an exponential function with a natural constant as the base; represents the clutter power; ρ represents the lag correlation coefficient; f dc represents the normalized clutter Doppler frequency; p0 represents the noise power, which is usually set to 1; I N represents the N-dimensional identity matrix (because the number of pulses is N, the vector z l , c l , p are all N-dimensional, and the covariance matrix R is N×N-dimensional).

[0111] In the simulation, it is assumed that the clutter obeys the compound-Gaussian distribution with inverse Gaussian texture (CG-IG), and the probability density function (PDF) of the texture is expressed as:

[0112]

[0113] Among them, ν represents the shape parameter and μ represents the scale parameter.

[0114] Four signal energy distributions are considered, as shown in Table 1. Signal model 1 has uniform energy distribution, signal models 2 and 3 represent uneven energy distribution, and most of the energy is concentrated in one distance unit, and signal model 4 represents an extreme case, where the signal energy of the distributed target is concentrated in one distance unit.

[0115] Table 1

[0116]

[0117] Experiment 1: The variation of PD with output SCR under different signal models 1 to 4 of the present invention, such as Figure 2 As shown. For the beauty of the picture, only the DT-MHM-DW and DT-MGM-DW detectors in the eighth formula are selected. Here, the number of pulses is set to 8, the number of distance units occupied by the target signal is 4, the number of training samples is 16, the clutter shape parameter is 0.9, the clutter scale parameter is 1.3, the covariance matrix lag correlation coefficient is 0.9, and the clutter power is 25dB.

[0118] Experiment 2: The variation of PD of the method of the present invention and the maximum eigenvalue detector with the output SCR; Figure 3 is signal mode 1; Figure 4 For signal mode 2; Figure 5 For signal mode 3; Figure 6 is signal mode 4; where the maximum eigenvalue detector expression of the distributed target is:

[0119]

[0120] in, Represents the HPD matrix When a=-1,1,0,-∞,∞, the detectors are called max-MHM-DW, max-MAM-DW, max-MGM-DW, max-MME-DW, and max-MEMD-DW, respectively.

[0121] In Experiment 2, the number of pulses was set to 8, the number of distance units occupied by the target signal was 4, the number of training samples was 16, the clutter shape parameter was 0.9, the clutter scale parameter was 1.3, the covariance matrix lag correlation coefficient was 0.9, and the clutter power was 25 dB.

[0122] Experiment 3: Figure 7-8 , the PD of the method of the present invention and the unwhitened detector varies with the output SCR; Figure 7 , the lag-one correlation coefficient of the covariance matrix is ​​0.9; Figure 8 , the covariance matrix lag-one correlation coefficient is 0.75; where the unwhitened detector of the distributed target is expressed as:

[0123]

[0124] in, is the HPD matrix The maximum eigenvalue of is the HPD matrix The nth eigenvalue of , and the HPD matrix By adding z l =[z 0,l ,...,z N-1,l ] T z in q,l , 0≤q≤N-1 instead Substitute the result into the sixth formula and substitute it into the fourth formula to calculate.

[0125] In Experiment 3, the number of pulses is set to 8, the number of distance units occupied by the target signal is 4, the number of training samples is 16, the clutter shape parameter is 0.9, the clutter scale parameter is 1.3, and the clutter power is 25 dB.

[0126] Experiment 4: The variation of clutter output SCR with the lag-one correlation coefficient ρ of the covariance matrix. The number of pulses is set to 8 and the clutter power is 10dB. Fig. 9 shown.

[0127] Fourth, conduct actual measurements using real sea clutter data;

[0128] The TFA17_014.03.mat dataset was used. This dataset was collected by the radar of the Council for Scientific and Industrial Research (CSIR) in South Africa. The clutter data follows the CG-IG distribution. Usually the shape parameter and scale parameter are unknown. Here, the shape parameter and scale parameter of the known CG-IG clutter are estimated to be 0.9 and 1.3 respectively through the moment estimation method. The number of pulses and the number of range cells in the TFA17_14.03.mat dataset are 133640 and 48 respectively. In order to be consistent with the simulation, the 17th to 20th range cells are used and the target signal s is added. l =β l p is used as the measured data, and the 8th to 15th and 22nd to 29th distance units form training samples. The variation of PD with output SCR of the method of the present invention and the unwhitened detector is shown in Fig.10 As shown in the figure, the number of pulses is set to 8, the number of distance units occupied by the target signal is 4, and the number of training samples is 16. In the actual measurement, SCR is defined as:

[0129]

[0130]

[0131] Among them, x i,j It is the element of the 100000×48 matrix formed by the first 100000 pulses and 48 distance cells of the dataset TFA17_014.03.mat.

[0132] The present invention also provides a radar distributed target detection system, and the specific technical solution is as follows:

[0133] An acquisition module, used for acquiring first radar data of distributed targets in a to-be-detected area, and performing a data whitening operation on the first radar data to obtain second radar data;

[0134] a calculation module, configured to determine a HPD matrix according to the second radar data, and determine an eigenvalue of the HPD matrix according to the HPD matrix;

[0135] The detector module is used to determine the first distributed target detector corresponding to the feature value through the correspondence between the feature value and the distributed target detector, obtain the detection result according to the first distributed target detector, and determine whether there is a distributed target according to the comparison result between the detection result and the preset value.

[0136] Based on the above solution, the present invention can also be improved as follows.

[0137] Further, the first radar data includes:

[0138] At least two of the target signal, clutter and Gaussian white noise.

[0139] Furthermore, the data whitening operation is specifically as follows:

[0140]

[0141] Furthermore, the corresponding relationship between the eigenvalue and the distributed target detector is:

[0142]

[0143] In the above embodiments, although the steps are numbered S1, S2, etc., these are only specific embodiments given by the present invention. Those skilled in the art may adjust the execution order of S1, S2, etc. according to actual conditions, which is also within the protection scope of the present invention. It can be understood that in some embodiments, some or all of the above embodiments may be included.

[0144] It should be noted that the beneficial effects of the radar distributed target detection system provided in the above embodiment are the same as the beneficial effects of the radar distributed target detection method provided in the above embodiment, which will not be described in detail here. In addition, when the system provided in the above embodiment realizes its functions, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the system can be divided into different functional modules according to actual conditions to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiment belong to the same concept, and the specific implementation process is detailed in the method embodiment, which will not be described in detail here.

[0145] like Fig.11 As shown, a computer device 300 according to an embodiment of the present invention includes a processor 320, the processor 320 is coupled to a memory 310, and the memory 310 stores at least one computer program 330, and the at least one computer program 330 is loaded and executed by the processor 320, so that the computer device 300 implements any of the above methods, specifically:

[0146] The computer device 300 may have relatively large differences due to different configurations or performances, and may include one or more processors 320 (Central Processing Units, CPU) and one or more memories 310, wherein at least one computer program 330 is stored in the one or more memories 310, and the at least one computer program 330 is loaded and executed by the one or more processors 320, so that the computer device 300 implements a radar distributed target detection method provided in the above embodiment. Of course, the computer device 300 may also have components such as a wired or wireless network interface, a keyboard, and an input and output interface for input and output, and the computer device 300 may also include other components for implementing device functions, which will not be repeated here.

[0147] A computer-readable storage medium according to an embodiment of the present invention stores at least one computer program, and the at least one computer program is loaded and executed by a processor so that a computer implements any one of the above methods.

[0148] Optionally, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0149] In an exemplary embodiment, a computer program product or a computer program is also provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs any of the above-mentioned radar distributed target detection methods.

[0150] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects and represent the definition of a specific order or sequence. The order of use of similar objects can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than the order shown or described.

[0151] Those skilled in the art know that the present invention can be implemented as a system, method or computer program product. Therefore, the present disclosure can be specifically implemented in the following forms, namely: it can be complete hardware, it can be complete software (including firmware, resident software, microcode, etc.), or it can be a combination of hardware and software, which is generally referred to as "circuit", "module" or "system" herein. In addition, in some embodiments, the present invention can also be implemented in the form of a computer program product in one or more computer-readable media, and the computer-readable medium contains computer-readable program code.

[0152] Any combination of one or more computer-readable media may be used. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, device, or device.

[0153] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A radar distributed target detection method, characterized in that: include: S1, acquiring first radar data of distributed targets in a to-be-detected area, and performing a data whitening operation on the first radar data to obtain second radar data; S2, determining an HPD matrix according to the second radar data, and determining an eigenvalue of the HPD matrix according to the HPD matrix, including: Substituting the second radar data into the fifth formula to obtain the correlation coefficient, the fifth formula is: Among them, r m,l represents the correlation coefficient, express The qth element of An expression representing the second radar data; The second radar data is averaged through the sixth formula to obtain an estimated value of the correlation coefficient, and the sixth formula is: in, represents the estimated value of the correlation coefficient, express The qth element of ; Substituting the estimated value of the correlation coefficient into the fourth formula to obtain the HPD matrix, the fourth formula is: in, represents the HPD matrix, Representation Statistics Expectations, (·) * denotes conjugation; The eigenvalue formula is used to extract the eigenvalue of the HPD matrix, and the eigenvalue is obtained. The characteristic value formula is: Among them, eig represents the HPD matrix Take the eigenvalue; S3, determining the first distributed target detector corresponding to the feature value through the correspondence between the feature value and the distributed target detector, obtaining a detection result according to the first distributed target detector, and determining whether a distributed target exists according to a comparison result between the detection result and a preset value.

2. A radar distributed target detection method according to claim 1, characterized in that: The first radar data includes: At least two of the target signal, clutter and Gaussian white noise.

3. A radar distributed target detection method according to claim 1, characterized in that: The data whitening operation is specifically as follows: in, represents the sample covariance matrix, K represents the number of units under test, L represents the number of training samples collected around the units under test, when l=1,...,K, z l Indicates the measured data, represents the whitened measured data. When l=K+1,...,K+L, z l represents the training sample data, represents the whitened training samples, Indicates z l Perform a conjugate transpose operation.

4. A radar distributed target detection method according to claim 1, characterized in that: The corresponding relationship between the eigenvalue and the distributed target detector is: Among them, t a,DT is the statistics of the detector, represents the HPD matrix, Indicates taking Maximum value, represents the eigenvalue of the HPD matrix, a represents the power parameter, N is the number of pulses in the CPI, and n=1,…,N represents the sequence number of the eigenvalue.

5. A radar distributed target detection system, characterized in that: include: An acquisition module, used for acquiring first radar data of distributed targets in a to-be-detected area, and performing a data whitening operation on the first radar data to obtain second radar data; A calculation module, used to determine an HPD matrix according to the second radar data, and determine an eigenvalue of the HPD matrix according to the HPD matrix, comprising: Substituting the second radar data into the fifth formula to obtain the correlation coefficient, the fifth formula is: Among them, r m,l represents the correlation coefficient, express The qth element of An expression representing the second radar data; The second radar data is averaged through the sixth formula to obtain an estimated value of the correlation coefficient, and the sixth formula is: in, represents the estimated value of the correlation coefficient, express The qth element of ; Substituting the estimated value of the correlation coefficient into the fourth formula to obtain the HPD matrix, the fourth formula is: in, represents the HPD matrix, Representation Statistics Expectations, (·) * denotes conjugation; The eigenvalue formula is used to extract the eigenvalue of the HPD matrix, and the eigenvalue is obtained. The characteristic value formula is: Among them, eig represents the HPD matrix Take the eigenvalue; The detector module is used to determine the first distributed target detector corresponding to the feature value through the correspondence between the feature value and the distributed target detector, obtain the detection result according to the first distributed target detector, and determine whether there is a distributed target according to the comparison result between the detection result and the preset value.

6. A radar distributed target detection system according to claim 5, characterized in that: The first radar data includes: At least two of the target signal, clutter and Gaussian white noise.

7. A radar distributed target detection system according to claim 6, characterized in that: The data whitening operation is specifically as follows: in, represents the sample covariance matrix, K represents the number of units under test, L represents the number of training samples collected around the units under test, when l = 1, ..., K, z l Indicates the measured data, represents the whitened measured data. When l=K+1,...,K+L, z l represents the training sample data, represents the whitened training samples, Indicates z l Perform a conjugate transpose operation.

8. A radar distributed target detection system according to claim 5, characterized in that: The corresponding relationship between the eigenvalue and the distributed target detector is: Among them, t a,DT is the statistics of the detector, represents the HPD matrix, Indicates taking Maximum value, represents the eigenvalue of the HPD matrix, a represents the power parameter, N is the number of pulses in the CPI, and n=1,…,N represents the sequence number of the eigenvalue.

9. A computer device, characterized in that: The computer device comprises a processor, the processor is coupled to a memory, at least one computer program is stored in the memory, and the at least one computer program is loaded and executed by the processor so that the computer device implements the method according to any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one computer program, and the at least one computer program is loaded and executed by a processor so that a computer implements the method according to any one of claims 1 to 4.

Images