Target detection method and device based on covariance matrix reconstruction under interference condition

Abstract

The application discloses a kind of robust target detection methods based on covariance matrix reconstruction under interference condition: first, training sample is used to form sampling covariance matrix;Then, the noise subspace is constructed using the sampling covariance matrix;Then, according to the angle information of target, the angle range to be used for covariance matrix reconstruction is determined, and the noise subspace is used to reconstruct the covariance matrix;Further, the reconstructed covariance matrix is used to construct detection statistics;According to the false alarm probability, the detection threshold is determined, and the size between detection statistics and detection threshold is compared, if detection statistics is greater than detection threshold, it is judged that target exists, otherwise, it is judged that target does not exist.The robust target detection method based on covariance matrix reconstruction under interference condition designed by the application has robust characteristics for interference angle mismatch, and when the interference angle information is inaccurate, the target can still be detected with a large detection probability.

Description

Technical Field

[0001] This invention belongs to the field of target detection technology, and more specifically, relates to a target detection method based on covariance matrix reconstruction under interference conditions, which is particularly suitable for multi-channel active phased array radar. Background Technology

[0002] Target detection is one of the most basic and essential functions of radar. However, interference poses a significant challenge to radar target detection, not only reducing the radar's detection probability but also increasing the false alarm probability. Adaptive beamforming followed by constant false alarm rate (CFAR) processing is a commonly used target detection method in the presence of interference. This method belongs to a step-by-step detection strategy, with beamforming and CFAR processing being the two core steps, which are independent of each other and cannot achieve optimal detection performance. Furthermore, if the positional relationship between the interference and the radar changes too rapidly, or if the interference exhibits positional perturbations, the angular information of the interference relative to the radar becomes uncertain. Traditional beamforming methods are not effective at suppressing interference, thus affecting target detection performance. Summary of the Invention

[0003] To overcome the limitations of detection performance and achieve robust target detection under the influence of target orientation disturbances, this invention proposes an integrated adaptive detection method based on the idea of ​​covariance matrix reconstruction, which can achieve target detection without the need for a separate interference suppression step.

[0004] To achieve the above objectives, this invention provides a target detection method based on covariance matrix reconstruction under interference conditions, comprising:

[0005] (1) Use training samples to form a sampling covariance matrix;

[0006] (2) Construct a noise subspace using the sampling covariance matrix;

[0007] (3) Determine the angle range and reconstruct the covariance matrix using the noise subspace;

[0008] (4) Construct the detection statistic using the reconstructed covariance matrix;

[0009] (5) Compare the detection statistic and the detection threshold. If the detection statistic is greater than the detection threshold, the target is determined to exist; otherwise, the target is determined to not exist.

[0010] In summary, the technical solutions conceived by this invention have the following beneficial effects compared with the prior art:

[0011] (1) The detector designed in this invention has robust characteristics against disturbances of the interference angle and can provide a higher target detection probability;

[0012] (2) The detector designed in this invention has constant false alarm characteristics and does not require additional constant false alarm processing steps. Attached Figure Description

[0013] Figure 1 This is a schematic diagram illustrating the principle of the target detection method based on covariance matrix reconstruction under interference conditions in this embodiment of the invention.

[0014] Figure 2 This is a schematic diagram of the target detection device based on covariance matrix reconstruction under interference conditions in an embodiment of the present invention. Detailed Implementation

[0015] 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 illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0016] For a phased array radar with N channels, assuming there are M interferences in space, each interference having an angle θ relative to the normal of the radar array. m m = 1, 2, ..., M; if a target exists, assume the angle of the target relative to the normal of the radar array is θ. t The radar received data can then be represented by an N×1 dimensional vector as follows:

[0017]

[0018] Among them, a t Represents the target amplitude, s(θ) t ) is the target guidance vector, and its expression is:

[0019]

[0020] d is the spacing between antenna elements, λ is the wavelength of the radar transmitted signal, and θ is the distance between antenna elements. t The direction and angle of the target are indicated by the superscript [·]. T Indicates transpose;

[0021] a m Let s(θ) be the amplitude of the m-th interference. m Let be the steering vector of the m-th disturbance, and its expression is:

[0022]

[0023] w represents the thermal noise in the data to be detected.

[0024] Based on formula (1), the interference plus noise covariance matrix of the data to be detected is:

[0025]

[0026] in, Let E[·] represent the power of the m-th interference, where E[·] represents the statistical expectation, |·| represents the absolute value, and the superscript [·] represents the absolute value. H This indicates the conjugate transpose operation. For thermal noise power, I N It is an N×N dimensional identity matrix.

[0027] In real-world environments, interference power Interference angle θ m and thermal noise power Since both the interference and noise components are unknown, the interference-noise covariance matrix R is also unknown. Therefore, a certain number of training samples are needed to estimate R. Assume there are L training samples containing only interference and noise components, denoted as the l-th training sample.

[0028]

[0029] Where a l,m Let m be the amplitude of the m-th perturbation in the l-th training sample, where l = 1, 2, ..., L, n l Let R be the thermal noise in the l-th training sample. Based on the training samples, the most commonly used estimator of R is the sampling covariance matrix.

[0030] However, for fast-moving interference, or for radars located on moving platforms, such as airborne radars, the azimuth angle θ of the interference is limited. m There may be differences in different training samples. In this case, if the sampling covariance matrix is ​​used... Using this as an estimate of the true covariance matrix R will introduce a large error, thereby reducing the target detection performance.

[0031] The purpose of this invention is to solve the problem of radar target detection when interference exists and the azimuth angle of the interference is disturbed. To achieve the above objective, as... Figure 1 As shown, this invention provides a robust target detection method based on covariance matrix reconstruction under interference conditions, comprising the following steps:

[0032] (1) Construct an N×N dimensional sampling covariance matrix using L training samples:

[0033]

[0034] Where x l For the l-th training sample, x l Containing only interference and noise components, denoted as a l,mLet m be the amplitude of the m-th perturbation in the l-th training sample, where l = 1, 2, ..., L, n l Let s(θ) be the thermal noise in the l-th training sample. m ) is the steering vector of the m-th interference, θ m Let x be the azimuth angle of the m-th interference; l , s(θ m ) and n l The dimension of each is N×1;

[0035] (2) Constructing the noise subspace using the sampling covariance matrix

[0036] U n =[u1,u2,…,u r (7)

[0037] Where u i Let be the i-th column of U, i = 1, 2, ..., r; and let U be the sampling covariance matrix. The characteristic matrix, i.e. Feature decomposition into U = [u1, u2, ..., u N ],=diag(λ1,λ2,…,λ N Let be a diagonal matrix, and let its eigenvalues ​​be arranged in descending order: λ1≥λ2≥…,λ N ;r is the smallest positive integer that satisfies the following relation:

[0038] (3) Determine the angle range and reconstruct the covariance matrix using the noise subspace:

[0039]

[0040] Where, θ k Let θ be the k-th component of the angle range Θ, and Θ is divided into 361 equal parts; the angle range Θ is determined in the following three cases: 1) when the target angle satisfies -88°≤θ t When ≤88°, the angle range is Θ=[-90°,θ low ]∪[θ up [90°], where it is assumed that the angle of the target relative to the normal direction of the radar array is θ. t ;2) When the target angle satisfies -90°≤θ t When <88°, the angle range is Θ=[θ up ,90°];3) When the target angle satisfies 88…<θ t When ≤90°, the angle range is Θ=[-90°,θ low ], where θ low =θ t -2°, θ up =θt +2°.

[0041] (4) Construct the detection statistic using the reconstructed covariance matrix:

[0042]

[0043] Wherein, s(θ) t The result is given in equation (2). Representation matrix The inverse of |·| represents the absolute value.

[0044] (5) Based on the false alarm probability, the detection threshold is determined by Monte Carlo simulation, and the detection statistic is compared with the detection threshold. If the detection statistic is greater than the detection threshold, the target is determined to exist; otherwise, the target is determined not to exist. The detection threshold is determined by Monte Carlo simulation.

[0045] Furthermore, such as Figure 2 As shown, this invention provides a target detection device based on covariance matrix reconstruction under interference conditions, including a sampling covariance matrix construction module, a noise subspace construction module, a covariance matrix reconstruction module, a detection statistic construction module, and a target decision module, wherein:

[0046] The sampling covariance matrix construction module is used to construct a 3D sampling covariance matrix using training samples;

[0047] The noise subspace construction module is used to construct a noise subspace using the sampling covariance matrix;

[0048] The covariance matrix reconstruction module is used to determine the angle range and reconstruct the covariance matrix using the noise subspace.

[0049] The detection statistic construction module is used to construct detection statistics using the reconstructed covariance matrix;

[0050] The target decision module is used to determine the detection threshold based on the false alarm probability and compare the detection statistic with the detection threshold. If the detection statistic is greater than the detection threshold, the target is determined to exist; otherwise, the target is determined to not exist.

[0051] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A target detection method based on covariance matrix reconstruction under interference conditions, characterized in that, include: (1) Constructing a sampling covariance matrix using training samples; including: constructing an N×N dimensional sampling covariance matrix using L training samples. Where the l-th training sample x l Containing only interference and noise components, denoted as a l,m Let m be the amplitude of the m-th perturbation in the l-th training sample, where l = 1, 2, ..., L, n l Let s(θ) be the thermal noise in the l-th training sample. m ) is the steering vector of the m-th interference, θ m Let x be the azimuth angle of the m-th interference; l , s(θ m ) and n l The dimension of each is N×1; the s(θ) m The expression for ) is d represents the spacing between antenna elements, λ represents the wavelength of the radar transmitted signal, and the superscript [·] represents the distance between the antenna elements. T Indicates transpose; (2) Construct the noise subspace using the sampling covariance matrix; the noise subspace is U n =[u1,u2,…,u r ], where u i Let be the i-th column of U, i = 1, 2, ..., r, and U be the sampling covariance matrix. The characteristic matrix, i.e. Feature decomposition into U = [u1, u2, ..., u N ], L=diag(λ1,λ2,…,λ N Let be a diagonal matrix, and let its eigenvalues ​​be arranged in descending order: λ1≥λ2≥…,λ N ;r is the smallest positive integer that satisfies the following relation: (3) Determine the angle range and reconstruct the covariance matrix using the noise subspace; the reconstructed covariance matrix is: Where, θ k Let θ be the k-th component of the angle range Θ, and Θ is divided into 361 equal parts; the angle range Θ is determined in the following three cases: 1) when the target angle satisfies -88°≤θ t When ≤88°, the angle range is Θ=[-90°,θ low ]∪[θ up [90°], where it is assumed that the angle of the target relative to the normal direction of the radar array is θ. t ;2) When the target angle satisfies -90°≤θ t When <88°, the angle range is Θ=[θ up ,90°];3) When the target angle satisfies 88°<θ t When ≤90°, the angle range is Θ=[-90°,θ low ], where θ low =θ t -2°, θ up =θ t +2°; (4) Construct the detection statistic using the reconstructed covariance matrix; the detection statistic is... Where s(θ) t ) is the target guidance vector, and its expression is: θ t The azimuth and angle of the target. For matrix The inverse of |·| represents the absolute value, and x is the data to be detected; (5) Compare the detection statistic and the detection threshold. If the detection statistic is greater than the detection threshold, the target is determined to exist; otherwise, the target is determined to not exist.

2. The target detection method based on covariance matrix reconstruction under interference conditions as described in claim 1, characterized in that, The detection threshold in step (5) is determined by Monte Carlo simulation.

3. A target detection device based on covariance matrix reconstruction under interference conditions, characterized in that, It includes a sampling covariance matrix construction module, a noise subspace construction module, a covariance matrix reconstruction module, a detection statistic construction module, and a target decision module, among which: The sampling covariance matrix construction module is used to construct a sampling covariance matrix using training samples; it includes: constructing an N×N dimensional sampling covariance matrix using L training samples. Where the l-th training sample x l Containing only interference and noise components, denoted as a l,m Let m be the amplitude of the m-th perturbation in the l-th training sample, where l = 1, 2, ..., L, n l Let s(θ) be the thermal noise in the l-th training sample. m ) is the steering vector of the m-th interference, θ m Let x be the azimuth angle of the m-th interference; l , s(θ m ) and n l The dimension of each is N×1; the s(θ) m The expression for ) is d represents the spacing between antenna elements, λ represents the wavelength of the radar transmitted signal, and the superscript [·] represents the distance between the antenna elements. T Indicates transpose; The noise subspace construction module is used to construct a noise subspace using the sampling covariance matrix; the noise subspace is U. n =[u1,u2,…,u r ], where u i Let be the i-th column of U, i = 1, 2, ..., r, and U be the sampling covariance matrix. The characteristic matrix, i.e. Feature decomposition into U = [u1, u2, ..., u N ], L=diag(λ1,λ2,…,λ N Let be a diagonal matrix, and let its eigenvalues ​​be arranged in descending order: λ1≥λ2≥…,λ N ;r is the smallest positive integer that satisfies the following relation: The covariance matrix reconstruction module is used to determine the angle range and reconstruct the covariance matrix using the noise subspace; the reconstructed covariance matrix is... Where, θ k Let θ be the k-th component of the angle range Θ, and Θ is divided into 361 equal parts; the angle range Θ is determined in the following three cases: 1) when the target angle satisfies -88°≤θ t When ≤88°, the angle range is Θ=[-90°,θ low ]∪[θ up [90°], where it is assumed that the angle of the target relative to the normal direction of the radar array is θ. t ;2) When the target angle satisfies -90°≤θ t When <88°, the angle range is Θ=[θ up ,90°];3) When the target angle satisfies 88°<θ t When ≤90°, the angle range is Θ=[-90°,θ low ], where θ low =θ t -2°, θ up =θ t +2°; The detection statistic construction module is used to construct a detection statistic using the reconstructed covariance matrix; the detection statistic is... Where s(θ) t ) is the target guidance vector, and its expression is: θ t The azimuth and angle of the target. For matrix The inverse of |·| represents the absolute value, and x is the data to be detected; The target decision module is used to determine the detection threshold based on the false alarm probability and compare the detection statistic with the detection threshold. If the detection statistic is greater than the detection threshold, the target is determined to exist; otherwise, the target is determined to not exist.

Images