A robust adaptive beamforming method and device based on prior information

Abstract

The application relates to the technical field of underwater acoustic signal processing, in particular to a robust adaptive beamforming method based on prior information. The method comprises the following steps: performing dimension reduction processing on a data covariance matrix according to a dimension reduction parameter in historical data to obtain a data dimension reduction covariance matrix and solving a data dimension reduction beamforming expression; deducing a preset value range of a loading quantity according to the data dimension reduction covariance matrix; setting an initial value of the loading quantity according to an optimal loading quantity parameter in historical data of a beam and solving a current optimal loading quantity; and if the current optimal loading quantity is within the preset value range, solving a steering vector of robust adaptive beamforming. The application sets a data dimension reduction order and an initial value in an iteration process by using a historical optimal data dimension reduction quantity and a historical optimal loading quantity, so that the convergence speed can be accelerated and the calculation complexity can be reduced. Meanwhile, the value range of the loading quantity is deduced to determine whether the diagonal loading quantity is effective, so that the iteration process is prevented from diverging.

Description

Technical Field

[0001] This application relates to the field of underwater acoustic signal processing technology, specifically to a robust adaptive beamforming method and apparatus based on prior information. Background Technology

[0002] Beamforming algorithms are a widely used technique in signal processing, primarily used to control the radiation patterns of antenna or sensor arrays to enhance signals in desired directions and suppress interference from other directions. Array distortion can degrade beamforming performance, especially in adaptive beamforming algorithms which are highly sensitive to distortion. To mitigate the impact of array distortion on adaptive beamforming algorithms, robust beamforming research is receiving increasing attention.

[0003] In related technologies, the robust adaptive beamforming algorithm (RCB) based on elliptic uncertainty sets constrains the error of the steering vector through ellipsoidal or spherical uncertainty sets, thus solving the steering vector error problem to a certain extent and increasing the anti-interference capability against steering vector error. However, RCB requires solving for the optimal loading amount in each direction, resulting in a huge computational burden, which is not suitable for real-time processing of large-scale arrays. To address the contradiction between robust performance and computational complexity in matrix spatial filtering, a data dimensionality reduction robust adaptive beamforming algorithm (RDRCB) has been proposed. RRDRCB reduces computational complexity while maintaining robustness, making it particularly suitable for processing real-time data in large arrays and dynamic environments. However, RRDRCB often uses Newton's iteration method or bisection method to solve for the optimal loading amount. Newton's iteration method has a fast convergence speed, but its disadvantage is that the loading amount is prone to invalid values. The bisection method has a slow convergence speed, but its advantage is robustness and reliability. Therefore, how to meet the real-time requirements of large-scale arrays for data processing without a significant decrease in signal-to-noise ratio gain compared to traditional methods has become an urgent problem for practitioners. Summary of the Invention

[0004] In related technologies, robust adaptive beamforming algorithms suffer from slow convergence speed and excessive computational complexity.

[0005] In a first aspect, embodiments of this application provide a robust adaptive beamforming method based on prior information, the robust adaptive beamforming method comprising: Obtain the data covariance matrix of the beam, and perform dimensionality reduction processing on the data covariance matrix based on the dimensionality reduction parameter in the historical best data of the beam to obtain the data dimensionality-reduced covariance matrix; Solve the beamforming expression for data dimensionality reduction based on the data dimensionality reduction covariance matrix; The preset range of loading values ​​is derived from the eigenvalues ​​and eigenvectors of the data dimensionality reduction covariance matrix. The initial value of the loading amount is set according to the optimal loading amount parameter in the historical data of the beam, and the beamforming expression after dimensionality reduction of the data is iteratively updated with the initial value until the current optimal loading amount is solved. If the current optimal loading amount is within the preset range, then the current optimal loading amount is used to solve for the driving vector formed by robust adaptive beamforming; If the current optimal loading amount is not within the preset range, reset the initial value of the loading amount and iterate again until the current optimal loading amount is within the preset range.

[0006] In conjunction with the first aspect, in one implementation, the step of reducing the data covariance matrix based on the dimensionality reduction parameter in the historical optimal data of the beam to obtain the data dimensionality-reduced covariance matrix includes: The dimensionality reduction order of the covariance matrix is ​​set based on the dimensionality reduction parameter in historical data; The initial dimensionality reduction matrix is ​​set based on the dimensionality reduction order and the data covariance matrix, and the optimal dimensionality reduction order of the data covariance matrix is ​​solved. Solve for the desired signal steering vector and constraint matrix based on the initial dimensionality reduction matrix.

[0007] In conjunction with the first aspect, in one implementation, setting the initial dimensionality reduction matrix based on the dimensionality reduction order and the data covariance matrix includes: Set the initial dimensionality reduction matrix D according to the formula:

[0008] In the formula, To estimate the steering vector of the desired signal, Let N be the data covariance matrix, and N be the dimensionality reduction order.

[0009] In conjunction with the first aspect, in one implementation, the step of solving the beamforming expression for data dimensionality reduction based on the data dimensionality reduction covariance matrix includes: The beamforming expression after data dimensionality reduction is generated based on the desired signal steering vector, constraint matrix, and data dimensionality reduction covariance matrix, and the Lagrange method is used to solve for the beamformer of the data dimensionality reduction.

[0010] In conjunction with the first aspect, in one implementation, the beamforming expression is:

[0011] In the formula, To guide the expected vector for dimensionality reduction, To reduce the dimension of the true guide vector, Let H be a dimension-reduced matrix, and R be the transpose of the matrix. y This is the covariance matrix for dimensionality reduction data.

[0012] In conjunction with the first aspect, in one embodiment, the beamformer employing the Lagrange method to solve for data dimensionality reduction includes: The beamforming expression function is solved using the Lagrange multiplier method, and the expression for the optimal value of the Lagrange factor is obtained.

[0013] In conjunction with the first aspect, in one implementation, the step of setting an initial value for the loading amount based on the optimal loading amount parameter in the historical beam data and iteratively updating the beamforming expression after dimensionality reduction of the data with this initial value until the current optimal loading amount is solved includes: The initial value of the Lagrange factor is set based on historical data, and the Lagrange factor is iteratively updated with this initial value until convergence, so as to obtain the current optimal loading amount of the Lagrange factor.

[0014] In conjunction with the first aspect, in one implementation, deriving the preset range of loading values ​​based on the eigenvalues ​​and eigenvectors of the data dimensionality reduction covariance matrix includes: The Lagrange factor is derived from the formula. The range of the optimal value:

[0015] In the formula, , for The One element, For eigenvalues, is the Lagrange factor, N is the order of dimensionality reduction, and H is the transpose.

[0016] In conjunction with the first aspect, in one implementation, the step of solving for the robust adaptive beamforming driving vector using the optimal loading amount includes: The dimensionality reduction steering vector is calculated using the obtained optimal loading amount, and the dimensionality reduction steering weighted vector is calculated based on the dimensionality reduction steering vector.

[0017] Secondly, embodiments of this application provide a robust adaptive beamforming method apparatus for implementing the robust adaptive beamforming method based on prior information as described in any of the above claims, characterized in that it includes: The data covariance matrix acquisition module is used to acquire the data covariance matrix of the beam. The dimension reduction processing module is connected to the data covariance matrix acquisition module and is used to perform dimension reduction processing on the data covariance matrix based on the dimension reduction parameter in the historical optimal data of the beam to obtain the data dimension-reduced covariance matrix. A beamforming calculation module, connected to the dimensionality reduction processing module, is used to solve the beamforming expression for data dimensionality reduction based on the data dimensionality reduction covariance matrix. The feature analysis module, connected to the beamforming calculation module, is used to derive the preset range of loading amount based on the eigenvalues ​​and eigenvectors of the data dimension reduction covariance matrix. The iterative update module includes a loading initial value setting unit and an iterative algorithm unit. The loading initial value setting unit is used to set the initial value of the loading amount according to the optimal loading amount parameter in the historical data of the beam. The iterative algorithm unit is used to iteratively update the beamforming expression after dimensionality reduction of the data with the initial value until the current optimal loading amount is solved. The judgment and generation module, connected to the iterative update module, includes a judgment unit and a driving vector generation unit. The judgment unit is used to determine whether the current optimal loading amount is within the preset value range. If it is within the preset value range, the driving vector generation unit uses the current optimal loading amount to solve for the driving vector formed by robust adaptive beamforming. The feedback processing module, connected to the judgment unit, is used to re-trigger the iterative algorithm unit to perform iterative updates when the current optimal loading amount is not within the preset value range, until the current optimal loading amount falls within the preset value range.

[0018] The beneficial effects of the technical solutions provided in this application include: This application's robust adaptive beamforming method utilizes historically optimal data dimensionality reduction and loading values ​​to set the data dimensionality reduction order and initial values ​​during the iteration process, thereby accelerating convergence and reducing computational complexity. Simultaneously, it derives the range of loading values ​​based on the eigenvalues ​​and eigenvectors of the data dimensionality reduction covariance matrix, using these values ​​to determine the effectiveness of the diagonal loading value and preventing divergence during the iteration process. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating a robust adaptive beamforming method in one embodiment of this application; Figure 2 A schematic diagram illustrating the iterative process of the RDRCB algorithm for solving the diagonal loading. Figure 3 This is a schematic diagram illustrating the iterative process of solving the diagonal loading amount using the robust adaptive beamforming method of this application; Figure 4 A comparative diagram showing the variation curves of the number of iterations for finding the optimal loading amount in various scenarios; Figure 5 This is a schematic diagram comparing the output signal-to-interference-plus-noise ratio (SNR) with the input SNR curve. Detailed Implementation

[0020] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0021] In related technologies, robust adaptive beamforming algorithms suffer from slow convergence speed and excessive computational complexity.

[0022] It is worth noting that, such as Figure 2 As shown, the RDRCB method is prone to divergence during the solution process, making it impossible to estimate the optimal loading amount.

[0023] In a first aspect, embodiments of this application provide a robust adaptive beamforming method based on prior information, the robust adaptive beamforming method comprising the following steps: Step S1: Obtain the data covariance matrix of the beam.

[0024] Specifically, calculate the data covariance matrix.

[0025] in, For the i-th snapshot array data, For the number of snapshots, For limited snapshot array data, This indicates the matrix transpose.

[0026] Step S2: Perform dimensionality reduction processing on the data covariance matrix based on the dimensionality reduction parameter in the historical best data of the beam to obtain the data dimensionality-reduced covariance matrix.

[0027] It is worth noting that in this application, the covariance matrix of the data is pre-processed by using the best historical data to reduce the dimensionality of the data covariance matrix to a suitable dimension for subsequent steps, thereby reducing the computational complexity.

[0028] Step S2 above includes: Step S2a: Set the dimensionality reduction order of the covariance matrix based on the dimensionality reduction parameter in the historical data.

[0029] Specifically, based on the optimal dimensionality reduction data from the previous batch of historical data. Set the initial dimensionality reduction order of the current covariance matrix. , For a moment.

[0030] Step S2b: Set the initial dimensionality reduction matrix based on the dimensionality reduction order and the data covariance matrix, and solve for the optimal dimensionality reduction order of the data covariance matrix.

[0031] Specifically, the initial dimensionality reduction matrix is ​​set according to the following formula. :

[0032] In the formula, To estimate the steering vector of the desired signal, Let N be the data covariance matrix, and N be the dimensionality reduction order.

[0033] Simultaneously solve for the current optimal dimensionality reduction. :

[0034] In the formula, For dimensionality reduction matrix column vectors, This represents the maximum boundary upper limit.

[0035] It is worth noting that this application uses the second-best dimensionality reduction as the initial value, which avoids multiple judgments and solutions. (Typically, the dimensionality reduction fluctuates between 10% and 15%). Step S2c: Solve for the desired signal steering vector and constraint matrix based on the initial dimensionality reduction matrix.

[0036] Specifically, solving for the desired signal steering vector and constraint matrix The expression is

[0037] in, E is the error identity matrix, and I is the identity matrix. The error between the preset guide vectors, The error between the preset guide vectors, It is the eigenvalue matrix.

[0038] Step S3: Solve the beamforming expression for data dimensionality reduction based on the data dimensionality reduction covariance matrix.

[0039] Step S3 specifically includes: Step S3a: Generate the beamforming expression after data dimensionality reduction based on the desired signal steering vector, constraint matrix, and data dimensionality reduction covariance matrix.

[0040] Specifically, the beamforming expression is as follows:

[0041] In the formula, b is the dimensionality reduction expectation steering vector. , To reduce the dimension of the true guide vector, F is the dimension reduction matrix, and H denotes the transpose. Ry is the covariance matrix of the reduced-dimensional data. For the desired guiding vector, This is the actual guide vector.

[0042] Step S3b: Solve for the beamformer by data dimensionality reduction using the Lagrange method.

[0043] Specifically, the beamforming expression function is solved using the Lagrange multiplier method.

[0044] Define a function:

[0045] In the formula, To reduce the dimensionality of the covariance matrix of the data, It is a Lagrange factor.

[0046] The above formula is correct. Taking the derivative and setting it to 0, we get:

[0047] Furthermore, Substitution From this, we can obtain the expression for the diagonal loading:

[0048] In the formula, for the matrix Perform eigenvalue decomposition , characteristic matrix It is a diagonal matrix, and , If the eigenvectors are the corresponding eigenvalues, then... , for The One element, These are the eigenvalues.

[0049] Step S4: Set the initial value of the loading amount based on the optimal loading amount parameter in the historical data of the beam, and use the initial value to iteratively update the beamforming expression after dimensionality reduction of the data until the current optimal loading amount is solved.

[0050] Step S4 specifically includes: Step S4a: Set the initial value of the Lagrange factor for iteration based on historical data, assuming the diagonal loading amount at time k. This is the initial value of the Lagrange multiplier. Set the convergence error. Set the iteration counter ; Step S4b: Iterate and update the Lagrange factor with the initial value until convergence to obtain the current optimal loading amount of the Lagrange factor.

[0051] Update the diagonal load, the process is as follows

[0052] Furthermore, repeat the above iterative update process until the solved Lagrange factor is equal to... The error between them is in the convergence error within, that is The current Lagrange factor is then taken as the current optimal loading amount.

[0053] It is worth noting that the iterative process is as follows: Figure 3 As shown, its iterative process is robust. And as... Figure 4 As shown, the curves represent the number of iterations for finding the optimal loading amount in various methods. The blue curve represents the robust adaptive beamforming method based on prior information in this application, the red curve represents the robust adaptive beamforming method based on the bisection method for data dimensionality reduction, and the black curve represents the robust adaptive beamforming method based on the Newton iteration method for data dimensionality reduction. It can be seen that the robust adaptive beamforming method based on prior information for data dimensionality reduction has approximately 2-3 iterations, which is significantly fewer than the other two algorithms.

[0054] Step S5: Derive the preset range of values ​​for the loading amount (which is the Lagrange factor) based on the eigenvalues ​​and eigenvectors of the data dimensionality reduction covariance matrix.

[0055] Specifically, the optimal value of the Lagrange factor (i.e., the optimal loading amount) is obtained based on the formula derivation. The range of values ​​for ) is:

[0056] In the formula, , for The One element, For eigenvalues, Here, N is the Lagrange multiplier, N is the order of dimensionality reduction, and H is the transpose. The desired signal is guided by a vector.

[0057] Step S6: Determine the current optimal loading amount of the Lagrange factor. Whether it is within the preset range. Among them, Case 1: If the current optimal loading amount is not within the preset value range, then re-execute steps S4 and S5, reset the initial value and iterate until the current optimal loading amount is within the preset value range.

[0058] Case 2: If the current optimal loading amount is within the preset range, then the current optimal loading amount is used to solve for the driving vector of the robust adaptive beamforming.

[0059] Specifically, the current optimal loading amount is obtained by utilizing the solved Lagrange factor. Calculate the dimension reduction steering vector :

[0060] Furthermore, the dimension reduction-guided weighted vector is calculated. :

[0061] In the formula, Reduce the dimensionality of the data using the covariance matrix.

[0062] It is worth noting that, such as Figure 5 As shown, in the absence of steering vector error, the signal-to-noise ratio gain curves of the RDRCB-PI algorithm are almost identical to those of the RDRCB algorithm, and the RDRCB-PI algorithm outperforms the Capon beamforming algorithm.

[0063] Secondly, this application provides a robust adaptive beamforming method apparatus for implementing the above-described robust adaptive beamforming method based on prior information, comprising: The system includes: a data covariance matrix acquisition module for acquiring the data covariance matrix of the beam; a dimensionality reduction processing module connected to the data covariance matrix acquisition module for reducing the dimensionality of the data covariance matrix based on the dimensionality reduction parameters in the beam's historical optimal data, to obtain a data dimensionality-reduced covariance matrix; a beamforming calculation module connected to the dimensionality reduction processing module for solving the beamforming expression for data dimensionality reduction based on the data dimensionality-reduced covariance matrix; a feature analysis module connected to the beamforming calculation module for deriving a preset range of loading values ​​based on the eigenvalues ​​and eigenvectors of the data dimensionality-reduced covariance matrix; and an iterative update module including a loading initial value setting unit and an iterative algorithm unit, wherein the loading initial value setting unit is used to calculate the loading value based on the optimal loading values ​​in the beam's historical data. The load parameter sets an initial value for the load. The iterative algorithm unit uses this initial value to iteratively update the beamforming expression after dimensionality reduction of the data until the current optimal load is found. A judgment and generation module, connected to the iterative update module, includes a judgment unit and a driving vector generation unit. The judgment unit determines whether the current optimal load is within a preset value range. If it is, the driving vector generation unit uses the current optimal load to solve for the robust adaptive beamforming driving vector. A feedback processing module, connected to the judgment unit, re-triggers the iterative algorithm unit to perform iterative updates when the current optimal load is not within the preset value range, until the current optimal load falls within the preset value range.

[0064] The functions of each module in the robust adaptive beamforming device based on prior information described above correspond to the steps in the embodiment of the robust adaptive beamforming method based on prior information described above, and their functions and implementation processes will not be described in detail here.

[0065] In summary, this paper optimizes the data dimensionality reduction robust adaptive beamforming algorithm by addressing its problems of slow convergence speed, long iteration count, and easy divergence in loading calculation. A new data dimensionality reduction robust adaptive beamforming algorithm based on prior information is invented, which has the advantages of high robustness, fast convergence speed, and short iteration count.

[0066] It should be noted that the sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0067] The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not indicate a sequence, nor do they limit "first," "second," and "third" to different types.

[0068] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.

[0069] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.

[0070] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.

[0071] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.

[0072] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A robust adaptive beamforming method based on prior information, characterized in that, The robust adaptive beamforming method includes: Obtain the data covariance matrix of the beam, and perform dimensionality reduction processing on the data covariance matrix based on the dimensionality reduction parameter in the historical data of the beam to obtain the data dimensionality-reduced covariance matrix; Solve the beamforming expression for data dimensionality reduction based on the data dimensionality reduction covariance matrix; The preset range of loading values ​​is derived from the eigenvalues ​​and eigenvectors of the data dimensionality reduction covariance matrix. The initial value of the loading amount is set according to the optimal loading amount parameter in the historical data of the beam, and the beamforming expression after dimensionality reduction of the data is iteratively updated with the initial value until the current optimal loading amount is solved. If the current optimal loading amount is within the preset range, then the current optimal loading amount is used to solve for the steering vector of robust adaptive beamforming; If the current optimal loading amount is not within the preset value range, the initial value of the loading amount is reset and iterative updates are performed again until the current optimal loading amount is within the preset value range. The preset range of loading values ​​derived from the eigenvalues ​​and eigenvectors of the data dimensionality reduction covariance matrix includes: The Lagrange factor is derived from the formula. The range of the optimal value: In the formula, , The desired signal steering vector, for The One element, For eigenvalues, is the Lagrange factor, N is the dimensionality reduction order, H is the transpose, max is the largest eigenvalue, and min is the smallest eigenvalue.

2. The robust adaptive beamforming method as described in claim 1, characterized in that, The step of reducing the data covariance matrix based on the dimensionality reduction parameter in the historical optimal data of the beam to obtain the data dimensionality-reduced covariance matrix includes: The dimensionality reduction order of the covariance matrix is ​​set based on the dimensionality reduction parameter in historical data; The initial dimensionality reduction matrix is ​​set based on the dimensionality reduction order and the data covariance matrix, and the optimal dimensionality reduction order of the data covariance matrix is ​​solved. Solve for the desired signal steering vector and constraint matrix based on the initial dimensionality reduction matrix.

3. The robust adaptive beamforming method as described in claim 2, characterized in that, The step of setting the initial dimensionality reduction matrix based on the dimensionality reduction order and the data covariance matrix includes: Set the initial dimensionality reduction matrix D according to the formula: In the formula, To estimate the steering vector of the desired signal, Let N be the data covariance matrix, and N be the dimensionality reduction order.

4. The robust adaptive beamforming method as described in claim 2, characterized in that, The step of solving the beamforming expression for data dimensionality reduction based on the data dimensionality reduction covariance matrix includes: The beamforming expression after data dimensionality reduction is generated based on the desired signal steering vector, constraint matrix, and data dimensionality reduction covariance matrix, and the Lagrange method is used to solve for the beamformer of the data dimensionality reduction.

5. The robust adaptive beamforming method as described in claim 4, characterized in that: The beamforming expression is: In the formula, To guide the desired dimension reduction vector, To reduce the dimension of the true guide vector, Let H be a dimension-reduced matrix, and R denote the transpose. y This is the covariance matrix for dimensionality reduction data.

6. The robust adaptive beamforming method as described in claim 4, characterized in that, The beamformer that uses the Lagrange method to solve for data dimensionality reduction includes: The beamforming expression function is solved using the Lagrange multiplier method, and the expression for the optimal value of the Lagrange factor is obtained.

7. The robust adaptive beamforming method as described in claim 6, characterized in that, The step of setting an initial value for the loading amount based on the optimal loading amount parameter in the historical beam data and iteratively updating the beamforming expression after dimensionality reduction of the data with this initial value until the current optimal loading amount is solved includes: The initial value of the Lagrange factor is set based on historical data, and the Lagrange factor is iteratively updated with this initial value until convergence, so as to obtain the current optimal loading amount of the Lagrange factor.

8. The robust adaptive beamforming method as described in claim 1, characterized in that, The method of solving the steering vector for robust adaptive beamforming using the optimal loading amount includes: The dimensionality reduction steering vector is calculated using the obtained optimal loading amount, and the dimensionality reduction steering weighted vector is calculated based on the dimensionality reduction steering vector.

9. A robust adaptive beamforming method apparatus for implementing the robust adaptive beamforming method based on prior information as described in claim 1, characterized in that, include: The data covariance matrix acquisition module is used to acquire the data covariance matrix of the beam. The dimension reduction processing module is connected to the data covariance matrix acquisition module and is used to perform dimension reduction processing on the data covariance matrix based on the dimension reduction parameter in the historical optimal data of the beam to obtain the data dimension-reduced covariance matrix. A beamforming calculation module, connected to the dimensionality reduction processing module, is used to solve the beamforming expression for data dimensionality reduction based on the data dimensionality reduction covariance matrix. The feature analysis module, connected to the beamforming calculation module, is used to derive the preset range of loading amount based on the eigenvalues ​​and eigenvectors of the data dimension reduction covariance matrix. The iterative update module includes a loading initial value setting unit and an iterative algorithm unit. The loading initial value setting unit is used to set the initial value of the loading amount according to the optimal loading amount parameter in the historical data of the beam. The iterative algorithm unit is used to iteratively update the beamforming expression after dimensionality reduction of the data with the initial value until the current optimal loading amount is solved. The judgment and generation module, connected to the iterative update module, includes a judgment unit and a driving vector generation unit. The judgment unit is used to determine whether the current optimal loading amount is within the preset value range. If it is within the preset value range, the driving vector generation unit uses the current optimal loading amount to solve for the steering vector of the robust adaptive beamforming. The feedback processing module, connected to the judgment unit, is used to re-trigger the iterative algorithm unit to perform iterative updates when the current optimal loading amount is not within the preset value range, until the current optimal loading amount falls within the preset value range.

Images