Robust adaptive beam forming method based on polarization sensitive array

Abstract

The invention relates to a robust adaptive beam forming method based on a polarization sensitive array. The method comprises the following steps: estimating a covariance matrix of interference and noise signals; obtaining the uncertain range of the airspace arrival angle of the expected signal, and calculating an accumulation matrix C according to the range and the airspace guide vector of the array, and representing the accumulation matrix C as follows: a pitch angle and an azimuth angle; performing eigendecomposition on the accumulation matrix C, and determining an eigenvector matrix corresponding to a main eigenvalue of the accumulation matrix C; calculating an optimal rotation vector based on the feature vector matrix and the interference and noise covariance matrix; calculating an optimal weight vector of beam forming based on the feature vector matrix and the optimal rotation vector; and performing beam forming on received data of the polarization sensitive array based on the optimal weight vector. According to the invention, the beam forming carried out when the angle of arrival is mismatched shows good robustness.

Description

Technical Field

[0001] This invention relates to the field of polarization-sensitive array signal processing technology, and more specifically to a robust adaptive beamforming method based on a polarization-sensitive array. Background Technology

[0002] Adaptive beamforming is the core of array signal processing, aiming to form a radiation pattern in space that points towards the desired signal and suppresses interference and noise by adjusting the weighting coefficients of the signals of each array element. Polarization-sensitive arrays (PSA) acquire multiple polarization components of electromagnetic waves at each element, utilizing not only the spatial information of the signal but also additional polarization information. This joint spatial-polarization processing capability allows PSAs to effectively distinguish and suppress interference and the desired signal even when their spatial characteristics are similar, leveraging their polarization differences. This results in superior anti-interference performance and signal reception capabilities compared to traditional scalar arrays.

[0003] In the field of PSA beamforming, the Minimum Variance Distortionless Response (MVDR) algorithm is a classic and widely used method. Its core idea is to minimize the total power of the array output (i.e., interference plus noise power) while ensuring distortion-free gain in the direction of the desired signal. However, the performance of the MVDR algorithm heavily relies on the precise knowledge of the desired signal's steering vector. In practical systems, such as active radar, the direction of arrival of the desired signal is usually preset based on the transmitted beam direction. Due to platform sway, beam pointing errors, or target micro-motion, a deviation inevitably exists between the preset direction and the actual direction of arrival, i.e., steering vector mismatch. This mismatched steering vector causes the MVDR beamformer to misjudge part of the desired signal as interference and suppress it, leading to a severe "signal self-cancellation" phenomenon and a sharp drop in the output signal-to-interference-plus-noise ratio (SINR).

[0004] To improve the robustness of beamforming algorithms under parameter mismatch conditions, various improvement schemes have been proposed in the industry. One mainstream method is based on diagonal loading (DL) technology, which improves the condition number of the matrix and widens the nulls by adding a suitable scalar identity matrix to the sampling covariance matrix, thereby tolerating errors to a certain extent. However, this method faces the dilemma of adaptively selecting the loading factor: if the loading is too small, the robustness is insufficient; if it is too large, the algorithm degenerates into conventional beamforming and loses its adaptive interference suppression capability. Another type of method, such as the power inversion (PI) algorithm, is suitable for scenarios where the desired signal is much weaker than the interference (such as satellite navigation). It minimizes the total output power by constraining the gain of the first channel of the array, thus eliminating the need to preset the direction of the desired signal. However, its radiation pattern lacks a clear main lobe, making it unsuitable for applications requiring enhanced desired signals (such as radar detection).

[0005] In recent years, researchers have proposed new approaches to address the challenge of imprecise spatial and polarization domain parameters. For example, Chinese invention patent CN118859099A discloses a robust beamforming method based on matrix reconstruction. This method first estimates the covariance matrix of the desired signal by performing spectral estimation and integral reconstruction on the received data covariance matrix within the spatial angular range where the desired signal may exist. Simultaneously, it reconstructs the interference plus noise covariance matrix in the interference region. Finally, it solves for the weight vector based on the maximum signal-to-noise ratio criterion. While this method does not rely on precisely known parameters, its matrix reconstruction process has high computational complexity, and its reconstruction performance is limited by the accuracy of the angle discretization density and the selection of the integration range. Another prior art (Chinese invention patent CN111585631A) focuses on achieving high-gain transmit beamforming with known precise parameters. Its robustness relies on accurate parameter estimation and is not suitable for scenarios with parameter mismatch.

[0006] In summary, when the desired signal angle of arrival is not precisely known (due to mismatch), existing polarization-sensitive array beamforming methods either face severe performance degradation, as is the case with traditional MVDR and diagonal loading methods, or have limitations such as computational complexity and dependence on parameter selection, as is the case with matrix reconstruction methods. Summary of the Invention

[0007] In view of the above-mentioned technical problems, this invention proposes a robust adaptive beamforming method based on a polarization-sensitive array. It does not require precise knowledge of the spatial domain information of the desired signal. Based on the approximate estimated spatial domain range, combined with the covariance matrix of the interference and noise signals, the optimal weight vector is obtained by maximizing the signal-to-interference-plus-noise ratio criterion for beamforming, and maintains excellent performance when the arrival angle of the desired signal is mismatched.

[0008] The technical solution to the technical problem of this invention is: a robust adaptive beamforming method based on a polarization-sensitive array, comprising the following steps:

[0009] Step S1: Estimate the covariance matrix of the interference plus noise signal. ;

[0010] Step S2: Obtain the spatial arrival angle uncertainty range of the desired signal. and according to the range and the array's spatial steering vector Calculate the cumulative matrix C, which is represented as: ,in, Indicates pitch angle, Indicates azimuth;

[0011] Step S3: Perform eigenvalue decomposition on the accumulated matrix C to determine the eigenvector matrix corresponding to its principal eigenvalues. ;

[0012] Step S4: Based on the feature vector matrix and the interference plus noise covariance matrix Calculate the optimal rotation vector ;

[0013] Step S5: Based on the feature vector matrix and the optimal rotation vector Calculate the optimal weight vector for beamforming. ;

[0014] Step S6: Based on the optimal weight vector Beamforming is performed on the received data from the polarization-sensitive array.

[0015] According to a technical solution of the present invention, in step S2, the uncertainty range of the spatial arrival angle of the desired signal is obtained. ,include:

[0016] Based on the main lobe width of the transmitted beam, the reference directions are obtained in the elevation angle θ domain and the azimuth angle φ domain, respectively. Determined as an intermediate value.

[0017] According to a technical solution of the present invention, the step of determining the eigenvector matrix corresponding to its principal eigenvalues ​​is... ,include:

[0018] Step S31: Perform eigenvalue decomposition on matrix C to obtain all eigenvalues;

[0019] Step S32: Arrange the feature values ​​in descending order and sum them sequentially until the ratio of the sum to the sum of all feature values ​​is greater than a preset threshold.

[0020] Step S33: Determine the eigenvalues ​​involved in the accumulation as principal eigenvalues, and form the eigenvector matrix by their corresponding eigenvectors. .

[0021] According to one technical solution of the present invention, the preset threshold is 90%.

[0022] According to one technical solution of the present invention, in step S4, the optimal rotation vector is calculated. Specifically, it includes:

[0023] Calculate matrix , is represented as: ;

[0024] Calculate matrix , is represented as: ;

[0025] Then the optimal rotation vector Represented as: ;

[0026] in, This represents the eigenvector corresponding to the largest eigenvalue of the matrix.

[0027] According to one technical solution of the present invention, the array element of the polarization sensitive array is a dual-polarization array element.

[0028] According to one technical solution of the present invention, the polarization sensitive array is arranged in an L-shape, a uniform linear array, or a conformal array.

[0029] According to one aspect of the present invention, a robust adaptive beamforming system based on a polarization-sensitive array is proposed, comprising:

[0030] The covariance matrix estimation module is used to estimate the covariance matrix of the interference plus noise signal. ;

[0031] The accumulation matrix calculation module is used to obtain the spatial arrival angle uncertainty range of the desired signal. and according to the range and the array's spatial steering vector Calculate the cumulative matrix C, which is represented as: ,in, Indicates pitch angle, Indicates azimuth;

[0032] The feature space acquisition module is used to perform eigenvalue decomposition on the accumulated matrix C to determine the eigenvector matrix corresponding to its principal eigenvalues. ;

[0033] The rotation vector solving module is used to solve for the eigenvector matrix. and the interference plus noise covariance matrix Calculate the optimal rotation vector ;

[0034] The weight vector calculation module is used to calculate the weight vector based on the feature vector matrix. and the optimal rotation vector Calculate the optimal weight vector for beamforming. ;

[0035] Beamforming output module, used for beamforming output based on the optimal weight vector Beamforming is performed on the received data from the polarization-sensitive array.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] This invention provides a robust adaptive beamforming method based on a polarization-sensitive array. By modeling the steering vector of the desired signal as residing within a characteristic subspace spanned by a spatial uncertainty range, rather than as a fixed vector, it avoids the signal self-cancellation problem caused by using a single mismatch steering vector. The method maintains high performance stability throughout the entire mismatch range, with the output SINR remaining close to its optimal value, demonstrating strong robustness.

[0038] The weight vector solution process of this invention only needs to utilize the spatial angular uncertainty range of the desired signal and the interference plus noise covariance matrix, without relying on the polarization parameters of the desired signal. This reduces the requirement for prior information, enabling the algorithm to work effectively in complex electromagnetic environments where polarization information is unknown or changing, and thus has wider applicability.

[0039] Compared to existing methods that reconstruct the covariance matrix based on spatial domain integration, this invention extracts the signal subspace by constructing a set of spatial steering vectors and performing a single eigenvalue decomposition, avoiding the cumbersome computations of dense sampling and multiple matrix inversions / operations in the angular domain. This method transforms the robustness problem into optimization within a reduced-dimensional feature subspace, achieving relatively lower computational complexity while maintaining high performance. Attached Figure Description

[0040] Figure 1 This is a flowchart illustrating a robust adaptive beamforming method based on a polarization-sensitive array in one embodiment of the present invention.

[0041] Figure 2 This is a schematic diagram of the composition of a robust adaptive beamforming system based on a polarization-sensitive array in one embodiment of the present invention;

[0042] Figure 3Output a graph showing the change of SINR as a function of SNR;

[0043] Figure 4 Output a graph showing the change in SINR as a function of snapshot number;

[0044] Figure 5 Output a graph showing the change in SINR as a function of the angle of arrival mismatch degree;

[0045] Figure 6 This is a flowchart illustrating a robust adaptive beamforming method based on a polarization-sensitive array, as shown in another embodiment of the present invention. Detailed Implementation

[0046] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the embodiments will be briefly described below. Obviously, the drawings described below are merely some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0047] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. The embodiments cannot be described in detail here, but the embodiments of the present invention are not limited to the following embodiments.

[0048] like Figure 1 and Figure 6 As shown, a robust adaptive beamforming method based on a polarization-sensitive array according to the present invention includes the following steps:

[0049] Step S1: Estimate the covariance matrix of the interference plus noise signal. ;

[0050] Step S2: Obtain the spatial arrival angle uncertainty range of the desired signal. and according to the range and the array's spatial steering vector Calculate the cumulative matrix C, which is represented as: ,in, Indicates pitch angle, Indicates azimuth;

[0051] Step S3: Perform eigenvalue decomposition on the accumulated matrix C to determine the eigenvector matrix corresponding to its principal eigenvalues. ,include:

[0052] Step S31: Perform eigenvalue decomposition on matrix C to obtain all eigenvalues;

[0053] Step S32: Arrange the feature values ​​in descending order and sum them sequentially until the ratio of the sum to the sum of all feature values ​​is greater than a preset threshold.

[0054] Step S33: Determine the eigenvalues ​​involved in the accumulation as principal eigenvalues, and form the eigenvector matrix by their corresponding eigenvectors. .

[0055] Preferably, the preset threshold is set to 90%;

[0056] Step S4: Based on the feature vector matrix and the interference plus noise covariance matrix Calculate the optimal rotation vector ;

[0057] Step S5: Based on the feature vector matrix and the optimal rotation vector Calculate the optimal weight vector for beamforming. ;

[0058] Step S6: Based on the optimal weight vector Beamforming is performed on the received data from the polarization-sensitive array.

[0059] In some embodiments of the present invention, the array elements of the polarization-sensitive array are dual-polarization array elements.

[0060] Taking a polarization-sensitive uniform L-shaped array as an example, the robust adaptive beamforming method based on a polarization-sensitive array mainly consists of three parts:

[0061] (1) Array signal model

[0062] The array consists of M pairs of orthogonal electric dipoles uniformly arranged along the z-axis and N pairs of orthogonal electric dipoles uniformly arranged along the y-axis, with an element spacing of d. Assume that K uncorrelated signals are incident on the polarization-sensitive L-shaped array, arranged sequentially from the 1st to the Mth element along the z-axis and from the 2nd to the Nth element along the y-axis to form the array manifold vector matrix. If the reference polarization element is located at the origin of the coordinate axes... For the incident signal of the k-th polarization array, This indicates the spatial angle of arrival of the signal. Indicating the polarization state of the signal, the received data from the array can be represented as:

[0063] (1)

[0064] in, yes 3D array manifold matrix, yes A dimensional incident signal vector, The mean is zero and the variance is... Complex Gaussian white noise.

[0065] Consider the phase difference of the incident signal with respect to the z-axis and y-axis array elements respectively:

[0066] (2)

[0067] (3)

[0068] The spatial steering vector of the array for the incident signal is:

[0069] (4)

[0070] The polarization steering vector is:

[0071] (5)

[0072] in, For the corresponding matrix of the array;

[0073] (6)

[0074] (7)

[0075] The array's steering vector can then be expressed as:

[0076] (8)

[0077] in This represents the Kronecker product. Furthermore, let... The received signal model of the array can be represented as:

[0078] (9)

[0079] (2) Setting the weight vector

[0080] Assume the range of the desired signal's spatial arrival angle. Given (e.g., based on the main lobe width of the transmitted beam), define matrix C:

[0081] (10)

[0082] By performing eigenvalue decomposition on matrix C, we can obtain:

[0083] (11)

[0084] in It is the eigenvector matrix. It is an eigenvalue matrix. It is the principal eigenvalue matrix. It is the eigenvector matrix corresponding to the principal eigenvalue matrix. Then the expected signal corresponds to... Must be located Zhang Cheng's space, that is

[0085] (12)

[0086] Therefore, it can be assumed that...

[0087] (13)

[0088] in Indicates to The transformation performed.

[0089] From a signal filtering perspective, the optimal weight vector should minimize the power of interference and noise after beamforming, while maximizing the output power of the desired signal. To achieve the suppression of interference and noise by the weight vector w and improve the output signal-to-interference-plus-noise ratio (SNR), the weight vector w can be solved by the following constrained optimization problem:

[0090] (14)

[0091] in, This indicates that the solution for variable w is to minimize the value of the function within the parentheses. Indicates constraint;

[0092] Similar to the method used to solve the MVDR optimization problem, the optimal weight vector can be obtained using the Lagrange multiplication method:

[0093] (15)

[0094] in, Although it cannot be accurately obtained The information, but combined with We can set the weight vector as

[0095] (16)

[0096] Where, vector satisfy .

[0097] (3) Solving for the optimal weight vector

[0098] The optimal weight vector can be obtained by maximizing the signal-to-interference-plus-noise ratio (SINR) of the beamformer output. ,Right now

[0099] (17)

[0100] Furthermore, due to the desired signal output power Under the premise that interference and noise are suppressed, it can be used Approximate estimation, therefore The following optimization problem can be solved.

[0101] (18)

[0102] Will Substituting into the above formula, corresponding vector The following optimization problem can be solved.

[0103] (19)

[0104] make , Then the above formula can be simply expressed as

[0105] (20)

[0106] Clearly, Q is a positive definite Hermitian matrix, therefore there exists a decomposition formula.

[0107] (twenty one)

[0108] Substituting the above equation into... From this, we can obtain

[0109] (twenty two)

[0110] make ,but ,and then

[0111] (twenty three)

[0112] Obviously If the matrix is ​​a positive definite Hermitian matrix, then This is a typical Rayleigh quotient expression. According to the properties of Rayleigh quotients, Maximum value that can be achieved It is a matrix The largest eigenvalue, then at this time Take the eigenvector corresponding to the largest eigenvalue. Therefore

[0113] (twenty four)

[0114] that is

[0115] (25)

[0116] Furthermore, we can obtain

[0117] (26)

[0118] Therefore, the optimal weight vector corresponding vector It is a matrix The eigenvector corresponding to the largest eigenvalue is

[0119] (27)

[0120] In the above formula This represents the eigenvector corresponding to the largest eigenvalue of the matrix. In this case, the weight vector is...

[0121] (28)

[0122] This weighting vector can exhibit superior beamforming performance when the polarization parameters are unknown and the desired signal angle of arrival is not precisely known.

[0123] The present invention will be described in detail below with a specific simulation example.

[0124] Simulation parameters: A uniform L-shaped polarized sensitive array is used, with 16 elements arranged along the z-axis and 16 elements arranged along the y-axis, with an element spacing of 0.5; the desired signal incident direction is... The direction of the interference signal incident is , The interference noise ratios are 0dB and 5dB, respectively; the polarization parameters of the desired signal and the interference signal are both... Assuming the polarization parameters are not mismatched, i.e. .

[0125] Simulation Condition 1: The desired signal-to-noise ratio ranges from -5dB to 20dB; the number of snapshots is N=1000; the pre-known direction of the desired signal is set to... , We conducted 200 independent simulation experiments, calculated the output signal-to-interference-plus-noise ratio (SINR), and compared it with traditional algorithms.

[0126] from Figure 3 The results show that, in the case of angle-of-arrival mismatch, except for those based on Apart from the MVDR method, the output signal-to-interference-plus-noise ratio of the other methods increases with the increase of the signal-to-noise ratio. The proposed robust beamforming method has the best performance and shows good robustness.

[0127] Simulation condition 2: The number of snapshots varies from 500 to 2000; the desired signal-to-noise ratio is set to 5dB; and the direction of the desired signal is known to be... , We conducted 200 independent simulation experiments, calculated the average output signal-to-interference-plus-noise ratio (SINR), and compared it with traditional algorithms.

[0128] from Figure 4 The results show that, under the condition of angle-of-arrival mismatch, the output signal-to-interference-plus-noise ratio of all algorithms does not change significantly with the increase of the number of snapshots. The proposed robust beamforming method has significant advantages over other algorithms and demonstrates good performance.

[0129] Simulation condition 3: The known mismatch range of the desired signal's elevation and azimuth angles is -5° to 5°; The desired signal-to-noise ratio (SNR) is set to 5 dB, and the number of snapshots N = 1000. 200 independent simulation experiments are conducted, the average output SNR is calculated, and compared with the results based on... The MVDR algorithm is compared with that of other algorithms.

[0130] from Figure 5 The results show that within a 5° mismatch range for both azimuth and elevation angles, the MVDR algorithm exhibits a significant drop, with a variation of up to 25dB. In contrast, the proposed robust beamforming method remains essentially unchanged, consistently maintaining a drop of around 17dB, which is similar to the MVDR algorithm results under the mismatched angle of arrival condition, demonstrating good robustness.

[0131] In summary, for beamforming in polarization-sensitive arrays where the desired signal angle of arrival (AOA) is inaccurate, this invention first establishes an optimal weight vector based on the uncertainty domain of the AOA, and then solves the problem using the interference plus noise covariance matrix based on the maximum signal-to-noise ratio criterion. Simulation results show that this algorithm exhibits good robustness in beamforming under AOA mismatch, outperforming other algorithms such as MVDR. Furthermore, this invention has the advantage of not requiring knowledge of polarization parameters.

[0132] According to one aspect of the present invention, a robust adaptive beamforming system based on a polarization-sensitive array is proposed, such as... Figure 2 As shown, it includes:

[0133] The covariance matrix estimation module is used to estimate the covariance matrix of the interference plus noise signal. ;

[0134] The accumulation matrix calculation module is used to obtain the spatial arrival angle uncertainty range of the desired signal. and according to the range and the array's spatial steering vector Calculate the cumulative matrix C, which is represented as: ,in, Indicates pitch angle, Indicates azimuth;

[0135] The feature space acquisition module is used to perform eigenvalue decomposition on the accumulated matrix C to determine the eigenvector matrix corresponding to its principal eigenvalues. ;

[0136] The rotation vector solving module is used to solve for the eigenvector matrix. and the interference plus noise covariance matrix Calculate the optimal rotation vector ;

[0137] The weight vector calculation module is used to calculate the weight vector based on the feature vector matrix. and the optimal rotation vector Calculate the optimal weight vector for beamforming. ;

[0138] Beamforming output module, used for beamforming output based on the optimal weight vector Beamforming is performed on the received data from the polarization-sensitive array.

[0139] The core idea of ​​the method described in this invention does not depend on a specific array geometry. While the specification uses an L-shaped array as an example, this method is also applicable to various polarization-sensitive array configurations, including uniform linear arrays, area arrays, circular arrays, and even conformal polarization arrays, demonstrating good versatility and scalability.

[0140] This invention presents a robust adaptive beamforming method based on a polarization-sensitive array, which addresses the performance degradation of beamforming algorithms caused directly or indirectly by steering vector mismatch when the desired signal angle of arrival (ROA) is inaccurate. The method first obtains a set of possible spatial steering matrices of the array based on the uncertainty domain of the ROA. Then, it calculates the covariance matrices of all possible spatial steering matrices and sums them to obtain an accumulation matrix. The rotation vector is obtained using the eigenspace of the accumulation matrix. Finally, the optimal weight vector is obtained based on the maximum signal-to-noise ratio (SNR) criterion. Compared to traditional methods such as MVDR, this method does not require precise knowledge of the spatial information of the desired signal. It can obtain the optimal weight vector for beamforming based on an approximately estimated spatial range, combined with the covariance matrix of the interference and noise signals, and using the maximum SNR criterion, maintaining excellent performance even when the ROA is mismatched.

[0141] Finally, it should be noted that the above description represents a preferred embodiment of the present invention. It should be pointed out that although preferred embodiments have been described, those skilled in the art, once they understand the basic inventive concept of the present invention, can make various improvements and modifications without departing from the principles described herein. These improvements and modifications should also be considered within the scope of protection of the present invention. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the embodiments of the present invention.

Claims

1. A robust adaptive beamforming method based on a polarization-sensitive array, characterized in that, Includes the following steps: Step S1: Estimate the covariance matrix of the interference plus noise signal. ; Step S2: Obtain the spatial arrival angle uncertainty range of the desired signal. and according to the range and the array's spatial steering vector Calculate the cumulative matrix C, which is represented as: ,in, Indicates pitch angle, Indicates azimuth; Step S3: Perform eigenvalue decomposition on the accumulated matrix C to determine the eigenvector matrix corresponding to its principal eigenvalues. ; Step S4: Based on the feature vector matrix and the interference plus noise covariance matrix Calculate the optimal rotation vector ; Step S5: Based on the feature vector matrix and the optimal rotation vector Calculate the optimal weight vector for beamforming. ; Step S6: Based on the optimal weight vector Beamforming is performed on the received data from the polarization-sensitive array.

2. The method according to claim 1, characterized in that, In step S2, the uncertainty range of the spatial arrival angle of the desired signal is obtained. ,include: Based on the main lobe width of the transmitted beam, the reference directions are obtained in the elevation angle θ domain and the azimuth angle φ domain, respectively. Determined as an intermediate value.

3. The method according to claim 1, characterized in that, The eigenvector matrix corresponding to its principal eigenvalue is determined. ,include: Step S31: Perform eigenvalue decomposition on matrix C to obtain all eigenvalues; Step S32: Arrange the feature values ​​in descending order and accumulate them sequentially until the ratio of the accumulated sum to the sum of all feature values ​​is greater than a preset threshold; Step S33: Determine the eigenvalues ​​involved in the accumulation as principal eigenvalues, and form the eigenvector matrix by their corresponding eigenvectors. .

4. The method according to claim 3, characterized in that, The preset threshold is 90%.

5. The method according to claim 4, characterized in that, In step S4, the optimal rotation vector is calculated. Specifically, it includes: Calculate matrix , is represented as: ; Calculate matrix , is represented as: ; Then the optimal rotation vector Represented as: ; in, This represents the eigenvector corresponding to the largest eigenvalue of the matrix.

6. The method according to claim 1, characterized in that, The array elements of the polarization-sensitive array are dual-polarization array elements.

7. The method according to claim 1, characterized in that, The polarization-sensitive array is arranged in one of the following ways: L-shaped, uniform linear array, or conformal array.

8. A robust adaptive beamforming system based on a polarization-sensitive array, characterized in that, include: The covariance matrix estimation module is used to estimate the covariance matrix of the interference plus noise signal. ; The accumulation matrix calculation module is used to obtain the spatial arrival angle uncertainty range of the desired signal. and according to the range and the array's spatial steering vector Calculate the cumulative matrix C, which is represented as: ,in, Indicates pitch angle, Indicates azimuth; The feature space acquisition module is used to perform eigenvalue decomposition on the accumulated matrix C to determine the eigenvector matrix corresponding to its principal eigenvalues. ; The rotation vector solving module is used to solve for the eigenvector matrix. and the interference plus noise covariance matrix Calculate the optimal rotation vector ; The weight vector calculation module is used to calculate the weight vector based on the feature vector matrix. and the optimal rotation vector Calculate the optimal weight vector for beamforming. ; Beamforming output module, used for beamforming output based on the optimal weight vector Beamforming is performed on the received data from the polarization-sensitive array.

Images