Robust intelligent detection of radar targets and adaptive suppression of jamming

Abstract

The present application relates to the technical field of wideband radar signal, and provides a radar target robust intelligent detection and interference adaptive suppression method and system, the method comprising: using the skew-symmetric characteristic of clutter covariance matrix, performing decorrelation transformation on the received data and parameter matrix to complete initial data preprocessing; respectively performing correlation calculation estimation under the condition of no target assumption and under the condition of target assumption, constructing the gradient detection statistics of distance extended target under the condition of known clutter skew-symmetric covariance matrix; constructing the detection statistics λ of the radar target robust intelligent detection and interference adaptive suppression method; determining the detection threshold T through theoretical calculation or Monte Carlo simulation according to the pre-set false alarm probability, comparing the detection statistics λ with the detection threshold T, and determining whether the current distance unit to be detected exists distance extended target according to the comparison result. The present application adopts analytical expression design, avoids complex iterative calculation process, and improves operation efficiency.

Description

Technical Field

[0001] This invention relates to the field of broadband radar signal technology, and in particular to a robust intelligent detection and adaptive interference suppression method and system for radar targets. Background Technology

[0002] With the increase in radar bandwidth, its range resolution has been further improved. Wideband radar is widely used in modern military and civilian fields such as anti-jamming, counter-reconnaissance, precision detection and imaging, high-precision tracking, and target recognition. Adaptive detection of range-extended targets in wideband radar has become one of the hot issues in the radar field. Unlike narrowband radar, where target echo signals usually occupy only one range resolution cell, the energy of a target scattering point in wideband radar may spread to adjacent range cells, presenting a "one-dimensional range image" and forming a range-extended target. If point target detection methods are still used to detect targets on echo signals for individual range cells and to estimate background clutter statistical characteristics by sampling from neighboring range cells, on the one hand, the energy of strong scattering points of range-extended targets is easily leaked into adjacent range cells, leading to signal contamination and further obscuring the target signal of a single range cell to be detected, resulting in poor performance of point target detection methods; on the other hand, in practical applications, radar detection faces complex electromagnetic environments, which may include natural or man-made interference sources such as electronic countermeasures signals or various civilian electromagnetic signals. In addition, the complex and variable environment in which the target is located enhances the non-uniformity of background clutter, and the number of training data for independent and identically distributed pure clutter is relatively limited. Compared with narrowband radar, this problem is particularly prominent in broadband radar target detection scenarios, making it difficult for existing range-extended target detection methods to achieve ideal detection results.

[0003] Furthermore, in commonly used rank-one signal target detection models, the target's steering vector is usually assumed to be a known, fixed vector. However, in practical applications, due to beam pointing errors and multipath phenomena, the target's steering vector may be mismatched. To address this issue, a subspace model can be used to model the target signal. In the subspace model, the signal is represented as the product of a known subspace matrix and an unknown coordinate matrix. If a subspace model is adopted for the target and interference signals based on a whole dataset consisting of test data and training data from multiple range cells to be detected, and detection statistics are constructed using the GLRT test criterion, a subspace GLRT detector for robust intelligent radar target detection (abbreviated as 2S-GLRT-HE) can be obtained. However, the construction process of the GLRT detector involves solving the maximum likelihood estimation of unknown parameters under both target and no-target assumptions, making the detector construction process complex and computationally complex. In addition, considering the difficulty in obtaining sufficient pure clutter training data in real-world environments, and the fact that the clutter covariance matrix of radar receivers using centrosymmetric linear arrays or centrosymmetric interval pulse trains has a special oblique symmetric structure. Utilizing oblique symmetric structural information can often improve the detection performance of detectors and reduce the amount of training data required. Considering oblique symmetric prior information, if a subspace modeling approach is adopted and the Rao test criterion is used to construct detection statistics, an oblique symmetric subspace Rao detector (abbreviated as P2S-Rao-HE) for robust intelligent radar target detection can be obtained. However, in practical implementation, the challenge of solving the Fisher information matrix is ​​encountered, and the detector construction process is also quite complex.

[0004] In situations with external interference and limited uniform training data, multi-channel broadband radar range-extended target detection faces the challenge of balancing computational complexity and detection performance. A key challenge is how to fully utilize oblique symmetric structural information, reduce the actual demand for training data, improve the estimation accuracy of unknown clutter covariance matrices, and construct a closed-form intelligent range-extended target detection method that effectively suppresses interference signals while maintaining constant false alarm rate (CFAR) characteristics, and achieves an effective balance between computational complexity and detection performance. This is crucial for enhancing the detection capabilities of broadband radar in complex interference environments and is one of the challenges faced by adaptive range-extended target detection in multi-channel broadband radar. Summary of the Invention

[0005] The purpose of this invention is to solve at least one technical problem in the background art and to provide a robust intelligent detection and adaptive interference suppression method and system for radar targets.

[0006] To achieve the above objectives, the present invention provides a robust intelligent radar target detection and adaptive interference suppression method, comprising:

[0007] Test data Z is collected from K distance cells to be tested, and training data Y is extracted from R adjacent reference cells. Using the oblique symmetry property of the clutter covariance matrix, decorrelation transformation is performed on the test data Z, clutter covariance matrix M, target signal coordinate matrix P, and interference signal coordinate matrix Q to complete the initial data preprocessing.

[0008] Without the objective assumption, solve for the maximum likelihood estimate of the interference signal coordinate matrix Q; with the objective assumption, solve for the objective parameter vector Θ. rp-1s Maximum likelihood estimation is performed, and the objective parameter vector Θ is obtained by using the complex Gaussian probability density function of the test data Z. rp-1s The partial derivative is calculated, and then the maximum likelihood estimate of the interference signal coordinate matrix Q obtained under the no-target assumption is substituted into the derivative result to obtain the calculation result. This calculation result and the maximum likelihood estimate of the target parameter vector under the target assumption are substituted into the range-extended target Gradient detection statistic to construct the range-extended target Gradient detection statistic under the known clutter oblique symmetric covariance matrix. Finally, the maximum likelihood estimate of the clutter oblique symmetric covariance matrix is ​​obtained based on the training data, and the maximum likelihood estimate of the clutter oblique symmetric covariance matrix is ​​substituted into the range-extended target Gradient detection statistic under the known clutter oblique symmetric covariance matrix, replacing the unknown clutter oblique symmetric covariance matrix, to construct the detection statistic λ of the radar target robust intelligent detection and interference adaptive suppression method.

[0009] The detection threshold T is determined by theoretical calculation or Monte Carlo simulation based on the pre-set false alarm probability. The detection statistic λ is compared with the detection threshold T, and the result of the comparison determines whether there is a range extension target in the current detection range cell.

[0010] According to one aspect of the present invention, collecting test data Z through K distance cells to be tested and extracting training data Y from R neighboring reference cells includes:

[0011] The radar system collects test data Z from K adjacent range cells. After appropriate sampling and organization, it forms an N×1 dimensional vector, represented as z. t ∈￡ N×1 ,t=1,2,...,K, where N represents the product of the number of antenna array elements and the number of pulses, Represents the set of N×1 dimensional complex matrices; the disturbance in the test data consists of clutter components and interference components, denoted as c0 and c1 respectively. t ∈￡ N×1 and j t ∈￡ N×1 ,t=1,2,...,K;Set the interference component j t The t = 1, 2, ..., K signals are modeled as deterministic subspace signals belonging to a known multi-rank subspace J ∈ £.N×q , represented as j t =Jq t , where q t ∈￡ q×1 t = 1, 2, ..., K represents the unknown complex coordinate vector of the interference signal; while clutter component c t The values ​​t = 1, 2, ..., K are independently and identically distributed across different distance cells, and follow a zero-mean circularly symmetric complex Gaussian distribution with covariance matrix M, i.e., c t ~CN(0 N×1 ,M); where M is an unknown positive Hermitian matrix; furthermore, the target signal s is set. t ,t=1,2,...,K is also modeled as a deterministic subspace signal, belonging to a known multi-rank subspace H∈￡ N×p , represented as s t =Hp t , where p t ∈￡ p×1 t = 1, 2, ..., K represents the unknown complex coordinate vector of the target signal;

[0012] For ease of derivation, we define an augmented matrix. To estimate the unknown clutter covariance matrix M, we assume there exists a set of training data containing only clutter components, typically collected from R range cells adjacent to the range cell to be measured, denoted as Y = [y1, y2, ..., y]. R ]∈￡ N×R , where y l =c l The ,l=1,2,...,R are also independent and identically distributed, and satisfy c l ~CN(0 N×1 M).

[0013] According to one aspect of the present invention, the oblique symmetry property of the clutter covariance matrix is ​​utilized to perform decorrelation transformation on the received data and parameter matrix to complete the initial data preprocessing, including:

[0014] Given the clutter covariance matrix M, the two-step Gradient test decision formula for a complex signal can be expressed as follows:

[0015]

[0016] Where, λ 2S-Gradient and T 2S-Gradient These represent the test statistic and threshold, respectively; the test data Z = [z1, z2, ..., z...]. K ]∈￡ N×K The target signal coordinate matrix P = [p1, p2, ..., p K ]∈￡ p×KThe interference signal coordinate matrix Q = [q1, q2, ..., q K ]∈￡ q×K Unknown parameter set Among them, the relevant parameter Θ r =vec(P)∈£ pK×1 Interference parameters This represents the maximum likelihood estimate of Θ under assumption H0. Represents Θ r Maximum likelihood estimation under the H1 assumption, Θ r0 Represents Θ r The truth value under the H0 hypothesis; superscript (·) T The vec() function represents the transpose of a matrix; it's important to note that under the H0 assumption, the target does not exist, therefore Θ... r0 =0 pK×1 f i (Z|iP,Q) is the conditional probability density function of the test data Z under hypothesis H0 or H1, expressed as:

[0017] f i (Z|iP,Q)={π N(K+R) |M| K+R} -1 ×exp[-tr(M -1 T i )],i=0,1 (2)

[0018] Among them, T i =(ZB i D)(ZB i D) H And define the augmented matrix B i =[iH,J] and D=[P T Q T ] T |·| represents the determinant of a square matrix, the tr(·) function takes the trace of the square matrix, the exp(·) function performs exponentiation, and the superscript (·)... H This represents the conjugate transpose; the condition p+q≤N must be satisfied to ensure that B=[H,J] is a column full-rank augmented matrix; furthermore, by utilizing the skew symmetry property of the clutter covariance matrix M, it can be derived that:

[0019] tr(M -1 T i )=tr[D N (M * ) -1 D N T i ] = tr(M -1 D N T i* D N ), i = 0, 1 (3)

[0020] Among them, the superscript (·) * The conjugate signifies conjugation; therefore, equation (2) can be rewritten as:

[0021] f i (Z|iP,Q)={π N(K+R) |M| K+R} -1 ×exp[-tr(M -1 T ip )],i=0,1 (4)

[0022] Among them, T ip =(T i +D N T i * D N ) / 2=(Z p -B i D p (Z) p -B i D p ) H ,and

[0023]

[0024] In the above formula, and These represent the real and imaginary parts, respectively.

[0025]

[0026] According to one aspect of the invention, in the absence of a target assumption, the maximum likelihood estimate of the interference signal coordinate matrix Q is solved; and in the presence of a target assumption, the target parameter vector Θ is solved. rp-1s Maximum likelihood estimation is performed, and the objective parameter vector Θ is obtained by using the complex Gaussian probability density function of the test data Z. rp-1sCalculate the partial derivative, then substitute the maximum likelihood estimate of the interference signal coordinate matrix Q obtained under the no-target assumption into the derivative to obtain the calculation result; substitute this calculation result and the maximum likelihood estimate of the target parameter vector under the target assumption into the range-extended target gradient detection statistic to construct the range-extended target gradient detection statistic under the known clutter oblique symmetric covariance matrix; finally, obtain the maximum likelihood estimate of the clutter oblique symmetric covariance matrix based on the training data, and substitute this maximum likelihood estimate into the range-extended target gradient detection statistic under the known clutter oblique symmetric covariance matrix, replacing the unknown clutter oblique symmetric covariance matrix, to construct the detection statistic λ of the radar target robust intelligent detection and interference adaptive suppression method, including:

[0027] Given the clutter covariance matrix M, the decision formula for the skew-symmetric two-step Gradient test can be reformulated as follows:

[0028]

[0029] in, express Maximum likelihood estimation under the H0 assumption, Θ rp-2s =vec(P p )∈￡ 2pK×1 Θ sp-2s =vec(Q p )∈￡ (2qK+1)×1 , Represents Θ rp-2s Under the H1 hypothesis, the conditional probability density function of the test data Z, based on the maximum likelihood estimation, can be expressed as:

[0030]

[0031] Under the assumption of H0, by differentiating equation (8) with respect to the interference signal coordinate matrix Q and setting the result to zero, we can obtain the maximum likelihood estimate of the interference signal coordinate matrix Q under the given clutter covariance matrix M:

[0032]

[0033] in, Under the H1 hypothesis, let the natural logarithm of equation (8) be Θ. rp-2s Taking the partial derivative, we can obtain:

[0034]

[0035] Substitute equation (9) into Further, there are:

[0036]

[0037] in, Indicates projection onto the matrix The orthogonal projection matrix of the column space of the orthogonal complement of the column space; next, let the natural logarithm of equation (8) be D. p Taking the partial derivative, we can obtain D. p The maximum likelihood estimate under the H1 assumption is:

[0038]

[0039] in, because Given the clutter covariance matrix M, we can obtain P p The maximum likelihood estimate under the H1 assumption is:

[0040]

[0041] Furthermore, we can obtain:

[0042]

[0043] Substituting equations (11) and (14) into equation (7), and then performing algebraic processing and simplification, we can obtain the Gradient test statistic given the clutter covariance matrix M as follows:

[0044]

[0045] in, Indicates projection onto the matrix The orthogonal projection matrix of the column space;

[0046] The maximum likelihood estimate of the unknown oblique-symmetric covariance matrix M is obtained using training data, which is exactly the same as the result in equation (5). Consistent; then, replace M in equation (15) with The final Gradient test statistic is:

[0047]

[0048] According to one aspect of the present invention, a detection threshold T is determined by theoretical calculation or Monte Carlo simulation based on a pre-set false alarm probability. The detection statistic λ is compared with the detection threshold T, and the presence of a range-extending target in the current detection range cell is determined based on the comparison result. This includes:

[0049] When λ≥T, it is determined that there is a distance extension target in the current detection distance unit, and the test data is marked as detected, excluding it from the subsequent sliding window training data update process;

[0050] When λ < T, it is determined that there is no range - extended target in the current range cell to be detected, and the current test data is incorporated into the training data set as a valid sample for updating the reference cell of the subsequent detection range cell.

[0051] To achieve the above object, the present invention further provides a radar target robust intelligent detection and interference adaptive suppression system, including:

[0052] A data pre - processing module collects test data Z through K range cells to be measured and extracts training data Y from R adjacent reference cells; using the skew - symmetric property of the clutter covariance matrix, it performs a decorrelation transformation on the test data Z, the clutter covariance matrix M, the target signal coordinate matrix P, and the interference signal coordinate matrix Q to complete the initial data pre - processing;

[0053] A detection statistic construction module, under the assumption of no target, solves the maximum likelihood estimate of the interference signal coordinate matrix Q; under the assumption of having a target, solves the maximum likelihood estimate of the target parameter vector Θ rp-1s of the maximum likelihood estimate, and takes the partial derivative of the target parameter vector Θ rp-1s using the complex Gaussian probability density function of the test data Z, and then substitutes the maximum likelihood estimate of the interference signal coordinate matrix Q obtained under the no - target assumption into the derivative result to obtain a calculation result; substitutes the calculation result and the maximum likelihood estimate of the target parameter vector under the target - having assumption into the range - extended target Gradient detection statistic to construct the range - extended target Gradient detection statistic under the condition of a known clutter skew - symmetric covariance matrix; finally, based on the training data, obtains the maximum likelihood estimate of the clutter skew - symmetric covariance matrix, and substitutes the maximum likelihood estimate of the clutter skew - symmetric covariance matrix into the range - extended target Gradient detection statistic under the condition of a known clutter skew - symmetric covariance matrix to replace the unknown clutter skew - symmetric covariance matrix therein, and constructs the detection statistic λ of the radar target robust intelligent detection and interference adaptive suppression method;

[0054] A target detection decision module determines the detection threshold T through theoretical calculation or Monte Carlo simulation according to the preset false - alarm probability, and determines whether there is a range - extended target in the current range cell to be detected according to the comparison result by comparing the detection statistic λ with the detection threshold T.

[0055] To achieve the above object, the present invention further provides an electronic device, including a processor, a memory, and a computer program stored on the memory and executable on the processor. When the computer program is executed by the processor, it implements the above - mentioned radar target robust intelligent detection and interference adaptive suppression method.

[0056] To achieve the above objectives, the present invention also provides a computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, it implements the radar target robust intelligent detection and interference adaptive suppression method as described above.

[0057] According to the solution of the present invention, the present invention can achieve the following beneficial effects:

[0058] 1) The radar target robust intelligent detection and interference adaptive suppression method adopts analytical expression design, which avoids complex iterative calculation process and significantly improves computational efficiency;

[0059] 2) Compared with methods such as Rao detection that require explicit calculation of complex Fisher information matrices, the method of this invention significantly reduces computational complexity, has a greater real-time advantage in engineering applications, and is easier to deploy and implement.

[0060] 3) By deeply exploring the skew symmetry characteristics of the clutter covariance matrix, the matrix estimation method was improved, which not only improved the estimation accuracy but also significantly reduced the requirements for training data, providing an efficient solution for distance-extended target detection under small sample conditions.

[0061] 4) For structured interference environments, the proposed detector can adaptively suppress interference signals of different intensities while maintaining robust detection performance against mismatch signals, significantly improving the detection reliability in complex electromagnetic environments.

[0062] 5) The method of the present invention can be effectively extended to some non-wideband radar detection scenarios, and is especially suitable for the following two typical applications: first, robust detection of large-sized targets by low / medium resolution radar; second, efficient detection of point target groups with the same speed of movement and dense spatial distribution (such as ship formations, aircraft formations, vehicle formations, etc.), which has important engineering application value. Attached Figure Description

[0063] Figure 1 A flowchart illustrating a radar target robust intelligent detection and interference adaptive suppression method according to an embodiment of the present invention is shown.

[0064] Figure 2 This is a comparison chart of the detection performance of the method of Embodiment 1 of the present invention and existing detection methods under a matching environment;

[0065] Figure 3 This is a comparison chart of the detection performance of the method of Embodiment 2 of the present invention and existing detection methods under mismatch conditions. Detailed Implementation

[0066] The invention will now be discussed with reference to exemplary embodiments. It should be understood that the described embodiments are merely intended to enable those skilled in the art to better understand and thus implement the invention, and are not intended to imply any limitation on the scope of the invention.

[0067] As used herein, the term "comprising" and its variations are to be interpreted as open-ended terms meaning "including but not limited to". The term "based on" is to be interpreted as "at least partially based on". The terms "one embodiment" and "an embodiment" are to be interpreted as "at least one embodiment".

[0068] Figure 1 The flowchart schematically illustrates a radar target robust intelligent detection and interference adaptive suppression method according to an embodiment of the present invention. Figure 1 As shown, in this embodiment, the radar target robust intelligent detection and interference adaptive suppression method includes:

[0069] Test data Z is collected from K distance cells to be tested, and training data Y is extracted from R adjacent reference cells. Taking advantage of the oblique symmetry of the clutter covariance matrix, decorrelation transformation is performed on the test data Z, clutter covariance matrix M, target signal coordinate matrix P, and interference signal coordinate matrix Q to complete the initial data preprocessing.

[0070] Without the objective assumption, solve for the maximum likelihood estimate of the interference signal coordinate matrix Q; with the objective assumption, solve for the objective parameter vector Θ. rp-1s Maximum likelihood estimation is performed, and the objective parameter vector Θ is obtained by using the complex Gaussian probability density function of the test data Z. rp-1s The partial derivative is calculated, and then the maximum likelihood estimate of the interference signal coordinate matrix Q obtained under the no-target assumption is substituted into the derivative result to obtain the calculation result. This calculation result and the maximum likelihood estimate of the target parameter vector under the target assumption are substituted into the range-extended target Gradient detection statistic to construct the range-extended target Gradient detection statistic under the known clutter oblique symmetric covariance matrix. Finally, the maximum likelihood estimate of the clutter oblique symmetric covariance matrix is ​​obtained based on the training data, and the maximum likelihood estimate of the clutter oblique symmetric covariance matrix is ​​substituted into the range-extended target Gradient detection statistic under the known clutter oblique symmetric covariance matrix, replacing the unknown clutter oblique symmetric covariance matrix, to construct the detection statistic λ of the radar target robust intelligent detection and interference adaptive suppression method.

[0071] The detection threshold T is determined by theoretical calculation or Monte Carlo simulation based on the pre-set false alarm probability. The detection statistic λ is compared with the detection threshold T, and the result of the comparison determines whether there is a range extension target in the current detection range cell.

[0072] Further, according to one embodiment of the present invention, collecting test data Z through K distance cells to be tested and extracting training data Y from R neighboring reference cells includes:

[0073] The radar system collects test data Z from K adjacent range cells. After appropriate sampling and organization, it forms an N×1 dimensional vector, represented as z. t ∈￡ N×1 ,t=1,2,...,K, where N represents the product of the number of antenna array elements and the number of pulses, Represents the set of N×1 dimensional complex matrices; the disturbance in the test data consists of clutter components and interference components, denoted as c0 and c1 respectively. t ∈￡ N×1 and j t ∈￡ N×1 ,t=1,2,...,K;Set the interference component j t The t = 1, 2, ..., K signals are modeled as deterministic subspace signals belonging to a known multi-rank subspace J ∈ £. N×q , represented as j t =Jq t , where q t ∈￡ q×1 t = 1, 2, ..., K represents the unknown complex coordinate vector of the interference signal; while clutter component c t The values ​​t = 1, 2, ..., K are independently and identically distributed across different distance cells, and follow a zero-mean circularly symmetric complex Gaussian distribution with covariance matrix M, i.e., c t ~CN(0 N×1 ,M); where M is an unknown positive Hermitian matrix; furthermore, the target signal s is set. t ,t=1,2,...,K is also modeled as a deterministic subspace signal, belonging to a known multi-rank subspace H∈￡ N×p , represented as s t =Hp t , where p t ∈￡ p×1 t = 1, 2, ..., K represents the unknown complex coordinate vector of the target signal;

[0074] For ease of derivation, we define an augmented matrix. To estimate the unknown clutter covariance matrix M, we assume there exists a set of training data containing only clutter components, typically collected from R range cells adjacent to the range cell to be measured, denoted as Y = [y1, y2, ..., y]. R ]∈￡ N×R , where y l =c l The ,l=1,2,...,R are also independent and identically distributed, and satisfy c l~CN(0 N×1 M).

[0075] Furthermore, according to one embodiment of the present invention, the oblique symmetry property of the clutter covariance matrix is ​​utilized to perform decorrelation transformation on the received data and parameter matrix to complete the initial data preprocessing, including:

[0076] Given the clutter covariance matrix M, the two-step Gradient test decision formula for a complex signal can be expressed as follows:

[0077]

[0078] Where, λ 2S-Gradient and T 2S-Gradient These represent the test statistic and threshold, respectively; the test data Z = [z1, z2, ..., z...]. K ]∈￡ N×K The target signal coordinate matrix P = [p1, p2, ..., p K ]∈￡ p×K The interference signal coordinate matrix Q = [q1, q2, ..., q K ]∈￡ q×K Unknown parameter set Among them, the relevant parameter Θ r =vec(P)∈£ pK×1 Interference parameters This represents the maximum likelihood estimate of Θ under assumption H0. Represents Θ r Maximum likelihood estimation under the H1 assumption, Θ r0 Represents Θ r The truth value under the H0 hypothesis; superscript (·) T The vec() function represents the transpose of a matrix; it's important to note that under the H0 assumption, the target does not exist, therefore Θ... r0 =0 pK×1 f i (Z|iP,Q) is the conditional probability density function of the test data Z under hypothesis H0 or H1, expressed as:

[0079] f i (Z|iP,Q)={π N(K+R) |M| K+R} -1 ×exp[-tr(M -1 T i )],i=0,1 (2)

[0080] Among them, T i =(ZB i D)(ZB i D) HAnd define the augmented matrix B i =[iH,J] and D=[P T Q T ] T |·| represents the determinant of a square matrix, the tr(·) function takes the trace of the square matrix, the exp(·) function performs exponentiation, and the superscript (·)... H This represents the conjugate transpose; the condition p+q≤N must be satisfied to ensure that B=[H,J] is a column full-rank augmented matrix; furthermore, by utilizing the skew symmetry property of the clutter covariance matrix M, it can be derived that:

[0081] tr(M -1 T i )=tr[D N (M * ) -1 D N T i ] = tr(M -1 D N T i * D N ), i = 0, 1 (3)

[0082] Among them, the superscript (·) * The conjugate signifies conjugation; therefore, equation (2) can be rewritten as:

[0083] f i (Z|iP,Q)={π N(K+R) |M| K+R} -1 ×exp[-tr(M -1 T ip )],i=0,1 (4)

[0084] Among them, T ip =(T i +D N T i * D N ) / 2=(Z p -B i D p (Z) p -B i D p ) H ,and

[0085]

[0086] In the above formula, and These represent the real and imaginary parts, respectively.

[0087]

[0088] Furthermore, according to one embodiment of the present invention, in the absence of a target assumption, the maximum likelihood estimate of the interference signal coordinate matrix Q is solved; in the presence of a target assumption, the target parameter vector Θ is solved. rp-1s Maximum likelihood estimation is performed, and the objective parameter vector Θ is obtained by using the complex Gaussian probability density function of the test data Z. rp-1s Calculate the partial derivative, then substitute the maximum likelihood estimate of the interference signal coordinate matrix Q obtained under the no-target assumption into the derivative to obtain the calculation result; substitute this calculation result and the maximum likelihood estimate of the target parameter vector under the target assumption into the range-extended target gradient detection statistic to construct the range-extended target gradient detection statistic under the known clutter oblique symmetric covariance matrix; finally, obtain the maximum likelihood estimate of the clutter oblique symmetric covariance matrix based on the training data, and substitute this maximum likelihood estimate into the range-extended target gradient detection statistic under the known clutter oblique symmetric covariance matrix, replacing the unknown clutter oblique symmetric covariance matrix, to construct the detection statistic λ of the radar target robust intelligent detection and interference adaptive suppression method, including:

[0089] Given the clutter covariance matrix M, the decision formula for the skew-symmetric two-step Gradient test can be reformulated as follows:

[0090]

[0091] in, express Maximum likelihood estimation under the H0 assumption, Θ rp-2s =vec(P p )∈￡ 2pK×1 Θ sp-2s =vec(Q p )∈￡ (2qK+1)×1 , Represents Θ rp-2s Under the H1 hypothesis, the conditional probability density function of the test data Z, based on the maximum likelihood estimation, can be expressed as:

[0092]

[0093] Under the assumption of H0, by differentiating equation (8) with respect to the interference signal coordinate matrix Q and setting the result to zero, we can obtain the maximum likelihood estimate of the interference signal coordinate matrix Q under the given clutter covariance matrix M:

[0094]

[0095] in, Under the H1 hypothesis, let the natural logarithm of equation (8) be Θ.rp-2s Taking the partial derivative, we can obtain:

[0096]

[0097] Substitute equation (9) into Further, there are:

[0098]

[0099] in, Indicates projection onto the matrix The orthogonal projection matrix of the column space of the orthogonal complement of the column space; next, let the natural logarithm of equation (8) be D. p Taking the partial derivative, we can obtain D. p The maximum likelihood estimate under the H1 assumption is:

[0100]

[0101] in, because Given the clutter covariance matrix M, we can obtain P p The maximum likelihood estimate under the H1 assumption is:

[0102]

[0103] Furthermore, we can obtain:

[0104]

[0105] Substituting equations (11) and (14) into equation (7), and then performing algebraic processing and simplification, we can obtain the Gradient test statistic given the clutter covariance matrix M as follows:

[0106]

[0107] in, Indicates projection onto the matrix The orthogonal projection matrix of the column space;

[0108] The maximum likelihood estimate of the unknown oblique-symmetric covariance matrix M is obtained using training data, which is exactly the same as the result in equation (5). Consistent; then, replace M in equation (15) with The final Gradient test statistic is:

[0109]

[0110] Furthermore, according to an embodiment of the present invention, to ensure the effectiveness of constant false alarm rate (CFAR) detection, the detection threshold T is determined through theoretical calculation or Monte Carlo simulation based on a preset false alarm probability. By comparing the detection statistic λ with the detection threshold T, it is determined whether there is a range extended target in the current range cell to be detected, including:

[0111] When λ≥T, it is determined that there is a range extended target in the current range cell to be detected. At the same time, the test data is marked as the detected state and excluded from the subsequent training data update process of the sliding window.

[0112] When λ<T, it is determined that there is no range extended target in the current range cell to be detected, and the current test data is used as a valid sample and incorporated into the training data set for updating the reference cell of the subsequent range cell to be detected.

[0113] According to the above solution of the present invention, for the problem that the existing wideband radar range extended target adaptive detector under a structured interference background is difficult to balance CFAR characteristics, detection performance and mismatch robustness, and considering the problem that it is difficult to obtain pure clutter training data due to actual clutter non-uniformity, how to fully exploit the structure information of the clutter covariance matrix, further reduce the requirement for the amount of training data, improve the estimation accuracy of the unknown clutter covariance matrix, and then construct a radar target robust intelligent detection and interference adaptive suppression method with a closed form, while ensuring CFAR characteristics, taking into account the intelligent anti-interference, detection performance and mismatch robustness and other multi-faceted requirements of the range extended target adaptive detection algorithm, and improving the adaptive detection performance of multi-channel wideband radar for weak targets in a complex interference environment.

[0114] According to the above solution of the present invention, the method of the present invention constructs a radar target robust intelligent detection and interference adaptive suppressor. It can be seen from Equation (16) that the proposed radar target robust intelligent detection and interference adaptive suppression method avoids the iterative calculation process in the traditional method by establishing an explicit detection statistic expression, significantly improving the operation efficiency. Compared with the P2S-Rao-HE detector, this method significantly reduces the implementation complexity by avoiding the calculation of the Fisher information matrix, while enhancing the engineering applicability. In addition, it is worth noting that compared with the 2S-GLRT-HE detector for range extended targets, the radar target robust intelligent detection and interference adaptive suppression method has a lower requirement for the amount of training data and stronger detection robustness for steering vector mismatch signals. Generally speaking, the radar target robust intelligent detection and interference adaptive suppression method of the present invention can effectively balance the algorithm calculation complexity, mismatch robustness and detection performance while maintaining CFAR characteristics.

[0115] Furthermore, to achieve the above objectives, the present invention also provides a radar target robust intelligent detection and interference adaptive suppression system, comprising:

[0116] The data preprocessing module collects test data Z from K distance cells to be measured and extracts training data Y from R adjacent reference cells. Utilizing the skew symmetry property of the clutter covariance matrix, it performs decorrelation transformation on the test data Z, the clutter covariance matrix M, the target signal coordinate matrix P, and the interference signal coordinate matrix Q to complete the initial data preprocessing.

[0117] The detection statistic construction module solves for the maximum likelihood estimate of the interference signal coordinate matrix Q when there is no objective assumption; and solves for the objective parameter vector Θ when there is an objective assumption. rp-1s Maximum likelihood estimation is performed, and the objective parameter vector Θ is obtained by using the complex Gaussian probability density function of the test data Z. rp-1s The partial derivative is calculated, and then the maximum likelihood estimate of the interference signal coordinate matrix Q obtained under the no-target assumption is substituted into the derivative result to obtain the calculation result. This calculation result and the maximum likelihood estimate of the target parameter vector under the target assumption are substituted into the range-extended target Gradient detection statistic to construct the range-extended target Gradient detection statistic under the known clutter oblique symmetric covariance matrix. Finally, the maximum likelihood estimate of the clutter oblique symmetric covariance matrix is ​​obtained based on the training data, and the maximum likelihood estimate of the clutter oblique symmetric covariance matrix is ​​substituted into the range-extended target Gradient detection statistic under the known clutter oblique symmetric covariance matrix, replacing the unknown clutter oblique symmetric covariance matrix, to construct the detection statistic λ of the radar target robust intelligent detection and interference adaptive suppression method.

[0118] The target detection and judgment module determines the detection threshold T based on the pre-set false alarm probability through theoretical calculation or Monte Carlo simulation. By comparing the detection statistic λ with the detection threshold T, it determines whether there is a range-extending target in the current detection range cell based on the comparison result.

[0119] The radar target robust intelligent detection and interference adaptive suppression system according to the present invention can realize the radar target robust intelligent detection and interference adaptive suppression method. The specific process steps are as described above and will not be repeated here.

[0120] Furthermore, to achieve the above objectives, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the radar target robust intelligent detection and interference adaptive suppression method as described above.

[0121] Furthermore, to achieve the above objectives, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the radar target robust intelligent detection and interference adaptive suppression method as described above.

[0122] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely one preferred embodiment of the invention and are only used to explain the invention. They do not limit the scope of protection of the invention. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0123] Example 1

[0124] Reference Figure 1 The radar target robust intelligent detection and interference adaptive suppression method in Example 1 consists of the following steps:

[0125] Step A1: Using a maritime detection radar, illuminate the sea area to be detected to obtain test data Z for K range units to be detected; use the test data Z to obtain oblique symmetric transformation test data Z according to equations (5) and (6). p ; Transform the oblique symmetry test data Z p Based on equation (9), Q is obtained under the H0 assumption. p Maximum likelihood estimation The test data Z under hypothesis H1 is obtained according to equation (11). p The complex Gaussian probability density function with respect to the target parameter vector Θ r The derivative of the given result is used to obtain Θ under the H1 assumption based on equation (14). rp Maximum likelihood estimation Based on the results obtained above, construct the range-extended target Gradient detection statistic under the condition of known clutter oblique symmetric covariance matrix according to equation (18).

[0126] It is worth noting that in step A1, considering the potential adverse effects of external interference on the adaptive detection of range-extended targets in the actual marine environment, the radar target robust intelligent detection and interference adaptive suppression method of this invention incorporates external interference into the detector design process and models the interference using subspace signals to reduce the potential mismatch effects of interference signals. For interference environments with subspace structure, the range-extended target Gradient intelligent detection method of this invention can effectively suppress interference signals of different intensities, exhibiting good intelligent anti-interference capabilities.

[0127] Step A2: Illuminate the targetless area surrounding the sea area to be detected with radar to obtain training data Y containing only pure sea clutter in R reference range cells. Then, use Equations (5) and (6) to obtain the maximum likelihood estimate of the clutter oblique-symmetric covariance matrix based on the training data Y. Will Substitute the range-extended target Gradient detection statistic obtained in step A1 into the unknown clutter oblique symmetric covariance matrix, and construct the detection statistic λ of the radar target robust intelligent detection and interference adaptive suppression method according to equation (19).

[0128] It is worth noting that, in step A2, compared with range-extended target detectors such as 2S-GLRT-HE and P2S-Rao-HE, the method of the present invention exhibits superior detection performance, especially under conditions of limited training data. Furthermore, the radar target robust intelligent detection and interference adaptive suppression method of the present invention has a closed-form expression. Compared with existing range-extended target adaptive detection methods, it maintains CFAR characteristics while achieving a reasonable balance between detection performance and computational complexity, thus improving the adaptive detection capability of multi-channel broadband radar for weak targets on the sea surface in complex electromagnetic environments.

[0129] Step A3: Set the detection threshold T according to the preset false alarm probability: Specifically, set the false alarm probability to P. fa According to the Monte Carlo method, based on the previously accumulated 100 / P fa The detection threshold T is calculated based on measured sea clutter data. Considering the difficulty in obtaining sea clutter data, if the actual amount of pure sea clutter measured data R is less than 100 / P... fa Then the missing 100 / P fa -R clutter data points can be obtained through simulation using a sea clutter simulation model, where the model parameters are reasonably estimated and set based on the obtained pure sea clutter measured data. Furthermore, the detection statistic λ is compared with the detection threshold T. If λ ≥ T, it is determined that there is a range-extending target in the current K range cells to be detected, and the test data is not used as training data for subsequent range cells to be detected; conversely, if λ < T, it is determined that there is no range-extending target in the current K range cells to be detected, and the test data is used as training data for subsequent range cells to be detected.

[0130] The performance comparison results of detectors under target-guided vector matching environment are shown in the appendix. Figure 2 The results show that, compared with existing range-extended target detectors such as 2S-GLRT-HE and P2S-Rao-HE, the method of this invention has better detection performance in a matched environment.

[0131] Example 2:

[0132] Reference Figure 1The radar target robust intelligent detection and interference adaptive suppression method in Example 2 consists of the following steps:

[0133] Step B1: Using a ground detection radar, illuminate the area to be detected to obtain test data Z for K range units to be detected; use the test data Z to obtain oblique symmetric transformation test data Z according to equations (5) and (6). p ; Transform the oblique symmetry test data Z p Based on equation (9), Q is obtained under the H0 assumption. p Maximum likelihood estimation The test data Z under hypothesis H1 is obtained according to equation (11). p The complex Gaussian probability density function with respect to the target parameter vector Θ r The derivative of the given result is used to obtain Θ under the H1 assumption based on equation (14). rp Maximum likelihood estimation Based on the results obtained above, construct the range-extended target Gradient detection statistic under the condition of known clutter oblique symmetric covariance matrix according to equation (18).

[0134] It is worth noting that in step B1, considering the potential adverse effects of external interference on the adaptive detection of range-extended targets in the actual marine environment, the radar target robust intelligent detection and interference adaptive suppression method of this invention incorporates external interference into the detector design process and models the interference using subspace signals to reduce the potential mismatch effects of interference signals. For interference environments with subspace structure, the range-extended target Gradient intelligent detection method of this invention can effectively suppress interference signals of different intensities, exhibiting good intelligent anti-interference capabilities.

[0135] Step B2: Illuminate the targetless area surrounding the area to be detected with radar to obtain training data Y containing only ground clutter in R reference range cells. Then, use Equations (5) and (6) to obtain the maximum likelihood estimate of the clutter oblique-symmetric covariance matrix based on the training data Y. Will Substitute the range-extended target Gradient detection statistic obtained in step B1 into the unknown clutter oblique symmetric covariance matrix, and construct the detection statistic λ of the radar target robust intelligent detection and interference adaptive suppression method according to equation (19).

[0136] It is worth noting that in step B2, compared with range-extended target detectors such as 2S-GLRT-HE and P2S-Rao-HE, the method of the present invention exhibits superior detection performance, especially under conditions of limited training data. Furthermore, the radar target robust intelligent detection and interference adaptive suppression method of the present invention has a closed-form expression. Compared with existing range-extended target adaptive detection methods, it maintains CFAR characteristics while achieving a reasonable balance between detection performance and computational complexity, thus improving the adaptive detection capability of multi-channel broadband radar for weak ground targets in complex electromagnetic environments.

[0137] Step B3: Set the detection threshold T according to the preset false alarm probability: Specifically, set the false alarm probability to P. fa According to the Monte Carlo method, based on the previously accumulated 100 / P fa The detection threshold T is calculated based on the measured ground clutter data. Considering the difficulty in acquiring ground clutter, if the actual amount of pure ground clutter measured data R is less than 100 / P... fa Then the missing 100 / P fa -R clutter data points can be obtained through simulation using a ground clutter simulation model, where the model parameters are reasonably estimated and set based on the obtained pure ground clutter measured data. Furthermore, the detection statistic λ is compared with the detection threshold T. If λ ≥ T, it is determined that the current K range cells to be detected contain range-extending targets, and the test data is not used as training data for subsequent range cells to be detected; conversely, if λ < T, it is determined that the current K range cells to be detected do not contain range-extending targets, and the test data is used as training data for subsequent range cells to be detected.

[0138] The results of the detector performance comparison under target-guided vector mismatch conditions are attached. Figure 3 The results show that, compared with existing range-extended target detectors such as 2S-GLRT-HE and P2S-Rao-HE, the method of this invention has better detection performance in mismatched environments.

[0139] Those skilled in the art will recognize that the modules and algorithm steps described in conjunction with the embodiments disclosed herein can be implemented using electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented 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 implementations should not be considered beyond the scope of this invention.

[0140] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described apparatus and equipment can be referred to the corresponding process in the foregoing method implementation, and will not be repeated here.

[0141] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0142] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the objectives of the embodiments of the present invention, depending on actual needs.

[0143] In addition, the functional modules in the embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0144] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the energy-saving signal transmission / reception methods of various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0145] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in this application.

[0146] It should be understood that the sequence number of each step in the invention and its embodiments does not absolutely imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

Claims

1. A robust intelligent radar target detection and adaptive interference suppression method, characterized in that, Including: Test data is collected using K distance units to be measured. Training data Y is extracted from the adjacent R reference cells; using the oblique symmetry property of the clutter covariance matrix, decorrelation transformation is performed on the test data Z, clutter covariance matrix M, target signal coordinate matrix P, and interference signal coordinate matrix Q to complete the initial data preprocessing. Without the objective assumption, solve for the maximum likelihood estimate of the interference signal coordinate matrix Q; with the objective assumption, solve for the objective parameter vector. Maximum likelihood estimation, and using test data The complex Gaussian probability density function with respect to the target parameter vector The partial derivative is calculated, and then the maximum likelihood estimate of the interference signal coordinate matrix Q obtained under the no-target assumption is substituted into the derivative to obtain the calculation result. This calculation result, along with the maximum likelihood estimate of the target parameter vector under the target assumption, is substituted into the range-extended target gradient detection statistic to construct the range-extended target gradient detection statistic under the known clutter oblique symmetric covariance matrix. Finally, the maximum likelihood estimate of the clutter oblique symmetric covariance matrix is ​​obtained based on the training data, and this maximum likelihood estimate is substituted into the range-extended target gradient detection statistic under the known clutter oblique symmetric covariance matrix, replacing the unknown clutter oblique symmetric covariance matrix, to construct the detection statistic for a robust intelligent radar target detection and adaptive interference suppression method. ; Determine the detection threshold T through theoretical calculation or Monte Carlo simulation based on a preset false alarm probability. By comparing the detection statistic λ with the detection threshold T, determine whether there is a range extended target in the currently to-be-detected range cell according to the comparison result.

2. The radar target robust intelligent detection and interference adaptive suppression method according to claim 1, characterized in that, Test data is collected using K distance units to be measured. And extract training data Y from the adjacent R reference units, including: The radar system collects test data from K adjacent range cells. After sampling and organization, it forms A dimensional vector is represented as Where N represents the product of the number of antenna array elements and the number of pulses, express A set of complex matrices of dimension ; the disturbances in the test data consist of clutter components and interference components, denoted as respectively. and , Set interference components Modeled as a deterministic subspace signal, belonging to a known multi-rank subspace. , represented as ,in , The unknown complex coordinate vector represents the interference signal; while the clutter component... They are independent and identically distributed across different distance cells, and follow a zero-mean circularly symmetric complex Gaussian distribution with a covariance matrix of M, i.e. Where M is an unknown positive Hermitian definite matrix; furthermore, the target signal is set. It is also modeled as a deterministic subspace signal, belonging to a known multi-rank subspace. , represented as ,in , The unknown complex coordinate vector representing the target signal; For ease of derivation, we define an augmented matrix. To estimate the unknown clutter covariance matrix M, we assume there exists a set of training data containing only clutter components, typically collected from R range cells adjacent to the range cell to be measured, denoted as M. ,in They are also independent and identically distributed, and satisfy... .

3. The radar target robust intelligent detection and interference adaptive suppression method according to claim 2, characterized in that, Utilize the skew-symmetric property of the clutter covariance matrix to perform a decorrelation transformation on the received data and parameter matrix, and complete the initial data preprocessing, including: Assume that the clutter covariance matrix M is known, then the two-step Gradient test decision formula for complex signals can be expressed as （1） in, and These represent the test statistic and threshold, respectively; test data. Target signal coordinate matrix Interference signal coordinate matrix Unknown parameter set Among them, the relevant parameters Interference parameters ; express Maximum likelihood estimation under assumption H0 express Maximum likelihood estimation under the H1 assumption express Truth value under the H0 hypothesis; superscript Indicates transpose. The function implements the vectorization of the matrix; it should be noted that under the H0 assumption, the target does not exist, therefore... , Let Z be the conditional probability density function of the test data under hypothesis H0 or H1, expressed as: （2） in, And define the augmented matrix. as well as ; The determinant of a square matrix. The function represents taking the trace of a square matrix. The function represents exponentiation, and the superscript indicates that the function is used for exponentiation. To represent the conjugate transpose; the following conditions must be met. To ensure It is a column full-rank augmented matrix; furthermore, by utilizing the skew symmetry property of the clutter covariance matrix M, it can be derived that: （3） Among them, superscript The conjugate signifies conjugation; therefore, equation (2) can be rewritten as: （4） in, ,and （5） In the above formula, and These represent the real and imaginary parts, respectively. （6）。 4. The radar target robust intelligent detection and interference adaptive suppression method according to claim 3, characterized in that, Without the objective assumption, solve for the maximum likelihood estimate of the interference signal coordinate matrix Q; with the objective assumption, solve for the objective parameter vector. Maximum likelihood estimation, and using test data The complex Gaussian probability density function with respect to the target parameter vector The partial derivative is calculated, and then the maximum likelihood estimate of the interference signal coordinate matrix Q obtained under the no-target assumption is substituted into the derivative to obtain the calculation result. This calculation result, along with the maximum likelihood estimate of the target parameter vector under the target assumption, is substituted into the range-extended target gradient detection statistic to construct the range-extended target gradient detection statistic under the known clutter oblique symmetric covariance matrix. Finally, the maximum likelihood estimate of the clutter oblique symmetric covariance matrix is ​​obtained based on the training data, and this maximum likelihood estimate is substituted into the range-extended target gradient detection statistic under the known clutter oblique symmetric covariance matrix, replacing the unknown clutter oblique symmetric covariance matrix, to construct the detection statistic for a robust intelligent radar target detection and adaptive interference suppression method. ,include: Assume that the clutter covariance matrix M is known, and the skew-symmetric two-step Gradient test decision formula can be reformulated as: （7） in, express Maximum likelihood estimation under the H0 assumption , , express Under the H1 hypothesis, the conditional probability density function of the test data Z, based on the maximum likelihood estimation, can be expressed as: （8） Under the hypothesis H0, let Equation (8) be differentiated with respect to the interference signal coordinate matrix Q, and set the result to zero, then the maximum likelihood estimate of the interference signal coordinate matrix Q under the condition of the given clutter covariance matrix M can be obtained: （9） in, , Under the H1 hypothesis, let the natural logarithm of equation (8) be a pair. Taking the partial derivative, we can obtain: （10） Substitute equation (9) into Furthermore: （11） in, ; , indicating projection onto the matrix The orthogonal projection matrix of the orthogonal complement of the column space; next, let the natural logarithm of equation (8) be the pair. Taking the partial derivative, we can obtain The maximum likelihood estimate under the H1 assumption is: （12） in, ;because Given the clutter covariance matrix M, we can obtain The maximum likelihood estimate under the H1 assumption is: （13） Furthermore, it can be obtained that: （14） Substitute Equation (11) and Equation (14) into Equation (7), and then perform algebraic processing and simplification, the Gradient test statistic under the condition of the given clutter covariance matrix M can be obtained as: （15） in, , indicating projection onto the matrix The orthogonal projection matrix of the column space; The maximum likelihood estimate of the unknown oblique-symmetric covariance matrix M is obtained using training data, which is exactly the same as the result in equation (5). Consistent; then, replace M in equation (15) with The final Gradient test statistic is: （16）。 5. The radar target robust intelligent detection and interference adaptive suppression method according to any one of claims 1-4, characterized in that, Determine the detection threshold T through theoretical calculation or Monte Carlo simulation based on a preset false alarm probability. By comparing the detection statistic λ with the detection threshold T, determine whether there is a range extended target in the currently to-be-detected range cell according to the comparison result, including: When λ ≥ T, determine that there is a range extended target in the currently to-be-detected range cell, and at the same time mark the test data as the detected state, excluding it from the subsequent training data update process of the sliding window. When λ < T, then determine that there is no range extended target in the currently to-be-detected range cell, and include the current test data as a valid sample in the training data set for the update of the reference cell of the subsequent to-be-detected range cell.

6. A radar target robust intelligent detection and interference adaptive suppression system, characterized in that, Including: The data preprocessing module collects test data from K distance units to be measured. Training data Y is extracted from the adjacent R reference cells; using the oblique symmetry property of the clutter covariance matrix, decorrelation transformation is performed on the test data Z, clutter covariance matrix M, target signal coordinate matrix P, and interference signal coordinate matrix Q to complete the initial data preprocessing. The detection statistic construction module solves for the maximum likelihood estimate of the interference signal coordinate matrix Q when there is no objective assumption; and solves for the objective parameter vector when there is an objective assumption. Maximum likelihood estimation, and using test data The complex Gaussian probability density function with respect to the target parameter vector The partial derivative is calculated, and then the maximum likelihood estimate of the interference signal coordinate matrix Q obtained under the no-target assumption is substituted into the derivative to obtain the calculation result. This calculation result, along with the maximum likelihood estimate of the target parameter vector under the target assumption, is substituted into the range-extended target gradient detection statistic to construct the range-extended target gradient detection statistic under the known clutter oblique symmetric covariance matrix. Finally, the maximum likelihood estimate of the clutter oblique symmetric covariance matrix is ​​obtained based on the training data, and this maximum likelihood estimate is substituted into the range-extended target gradient detection statistic under the known clutter oblique symmetric covariance matrix, replacing the unknown clutter oblique symmetric covariance matrix, to construct the detection statistic for a robust intelligent radar target detection and adaptive interference suppression method. ; A target detection determination module that determines the detection threshold T through theoretical calculation or Monte Carlo simulation based on a preset false alarm probability. By comparing the detection statistic λ with the detection threshold T, determine whether there is a range extended target in the currently to-be-detected range cell according to the comparison result.

7. An electronic device, characterized in that, Including a processor, a memory, and a computer program stored on the memory and executable on the processor. When the computer program is executed by the processor, it implements the radar target robust intelligent detection and interference adaptive suppression method as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium. When the computer program is executed by the processor, it implements the radar target robust intelligent detection and interference adaptive suppression method as described in any one of claims 1-5.

Images