Method, device, equipment and medium for estimating angle of arrival of underwater acoustic array based on adaptive kernel width mixed correlation entropy

Abstract

The present application discloses a method, device, equipment and medium for estimating the arrival angle of an underwater acoustic array based on adaptive kernel-width mixed correlation entropy, and relates to the technical field of underwater acoustic parameter estimation, including: optimizing the initial data matrix based on a preset mixed correlation entropy noise filtering method and an adaptive kernel width value to obtain a target data matrix; performing array arrival angle estimation on the target data matrix to obtain spatial spectrum information for characterizing the target orientation; when the estimated angle difference between any signal sources in the spatial spectrum information is greater than the preset angle difference, determining that the spatial spectrum information is a valid estimate. The present application can effectively improve the accuracy of underwater acoustic array DOA estimation.

Description

Technical Field

[0001] The present application relates to the technical field of underwater acoustic parameter estimation, and in particular to an underwater acoustic array arrival angle estimation method, device, equipment and medium based on adaptive kernel-width mixed correlation entropy. Background Art

[0002] The traditional underwater acoustic array Direction of Arrival (DOA) estimation method does not take into account the non-Gaussian characteristics of the underwater environment, which will reduce the estimation accuracy.

[0003] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0004] The main purpose of the present application is to provide a method, device, equipment and medium for estimating the angle of arrival of an underwater acoustic array based on adaptive kernel-width mixed correlation entropy, aiming to solve the technical problem that the traditional underwater acoustic array DOA estimation method does not take into account the non-Gaussian characteristics of the underwater environment, resulting in a decrease in estimation accuracy.

[0005] To achieve the above object, the present application proposes a method for estimating the angle of arrival of an underwater acoustic array based on adaptive kernel-width mixed correlation entropy. The method for estimating the angle of arrival of an underwater acoustic array based on adaptive kernel-width mixed correlation entropy includes:

[0006] Based on the preset mixed correlation entropy noise filtering method and the adaptive kernel width value, the initial data matrix is ​​optimized to obtain the target data matrix;

[0007] Performing array arrival angle estimation on the target data matrix to obtain spatial spectrum information for characterizing the target orientation;

[0008] When the estimated angle differences between any signal sources in the spatial spectrum information are greater than a preset angle difference, the spatial spectrum information is determined to be a valid estimate.

[0009] In one embodiment, the step of optimizing the initial data matrix based on the preset hybrid correlation entropy noise filtering method and the adaptive kernel width value to obtain the target data matrix includes:

[0010] S101, receiving input data, wherein the input data includes an initial data matrix, a maximum number of iterations, and a maximum allowable relative error;

[0011] S102, initializing a first random matrix, a second random matrix and an iteration parameter according to the initial data matrix, wherein the first random matrix is ​​a full column rank random matrix, and the second random matrix is ​​a full row rank random matrix;

[0012] S103, calculating a first high-order matrix of the first random matrix and calculating a second high-order matrix of the second random matrix based on a preset hybrid correlation entropy noise filtering method and an adaptive kernel width value;

[0013] S104. Update the iteration parameters according to the first random matrix, the first high-order matrix, the second random matrix and the second high-order matrix, and use the first high-order matrix as a new first random matrix and the second high-order matrix as a new second random matrix to repeat steps S103-S104 until the current iteration parameters meet the convergence conditions, use the current first high-order matrix and the current second high-order matrix as the first optimal matrix and the second optimal matrix respectively, and determine the target data matrix according to the first optimal matrix and the second optimal matrix, wherein the iteration parameters include the number of iterations and the iteration error, and the convergence condition is that the current number of iterations is greater than the maximum number of iterations or the current iteration error is greater than the maximum allowable relative error.

[0014] In one embodiment, the step of calculating a first high-order matrix of the first random matrix and calculating a second high-order matrix of the second random matrix based on a preset hybrid correlation entropy noise filtering method and an adaptive kernel width value includes:

[0015] Based on the preset mixed correlation entropy noise filtering method, a cost function of a maximum mixed correlation entropy criterion is constructed according to a Gaussian function and an adaptive kernel width value;

[0016] Based on the cost function, a first high-order matrix of the first random matrix is ​​calculated through an alternating optimization strategy, and a second high-order matrix of the second random matrix is ​​calculated.

[0017] In one embodiment, the step of constructing a cost function of the maximum mixed correlation entropy criterion according to the Gaussian function and the adaptive kernel width value includes:

[0018] Determine the residual fitting error value of the current iteration;

[0019] Determining an adaptive kernel width value based on the residual fitting error value, the proportionality coefficient and the preset kernel width;

[0020] A cost function of a maximum mixed correlation entropy criterion is constructed according to the Gaussian function and the adaptive kernel width value.

[0021] In one embodiment, the step of calculating the first high-order matrix of the first random matrix by an alternating optimization strategy based on the cost function, and calculating the second high-order matrix of the second random matrix comprises:

[0022] Based on the cost function and the alternating optimization strategy, iteratively determine a first independent sub-item according to the first random matrix, and determine a first high-order matrix according to the first independent sub-item;

[0023] Based on the cost function and the alternating optimization strategy, a second independent sub-item is iteratively determined according to the second random matrix, and a first high-order matrix is ​​determined according to the second independent sub-item.

[0024] In one embodiment, the step of updating the iteration parameter according to the first random matrix, the first high-order matrix, the second random matrix and the second high-order matrix comprises:

[0025] Determining a first cost function value based on the cost function, the first random matrix, and the second random matrix;

[0026] Determining a second cost function value based on the cost function, the first high-order matrix, and the second high-order matrix;

[0027] Determining a relative error according to the first cost function value and the second cost function value;

[0028] The iteration error is updated according to the relative error, and the number of iterations is updated.

[0029] In one embodiment, the step of performing array arrival angle estimation on the target data matrix to obtain spatial spectrum information for characterizing the target orientation includes:

[0030] Determining a covariance matrix of the target data matrix;

[0031] Performing eigenvalue decomposition on the covariance matrix to obtain an eigenvector matrix;

[0032] Determine a signal subspace according to the eigenvector matrix, and determine a noise subspace according to the signal subspace;

[0033] Based on the noise subspace, the array arrival angle of the target data matrix is ​​estimated to obtain spatial spectrum information for characterizing the target orientation.

[0034] In addition, to achieve the above purpose, the present application also proposes an underwater acoustic array arrival angle estimation device based on adaptive kernel width mixed correlation entropy, and the underwater acoustic array arrival angle estimation device based on adaptive kernel width mixed correlation entropy includes:

[0035] An optimization module is used to optimize the initial data matrix based on a preset mixed correlation entropy noise filtering method and an adaptive kernel width value to obtain a target data matrix;

[0036] The output module is used to estimate the array arrival angle of the target data matrix to obtain spatial spectrum information for characterizing the target orientation.

[0037] The determination module is configured to determine that the spatial spectrum information is a valid estimation when the estimated angle differences between any signal sources in the spatial spectrum information are greater than a preset angle difference.

[0038] In addition, to achieve the above-mentioned objectives, the present application also proposes a device for estimating the angle of arrival of an underwater acoustic array based on adaptive kernel-width mixed correlation entropy, the device comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the steps of the method for estimating the angle of arrival of an underwater acoustic array based on adaptive kernel-width mixed correlation entropy as described above.

[0039] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the underwater acoustic array arrival angle estimation method based on adaptive kernel width mixed correlation entropy as described above are implemented.

[0040] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the underwater acoustic array arrival angle estimation method based on adaptive kernel width mixed correlation entropy as described above.

[0041] One or more technical solutions proposed in this application have at least the following technical effects:

[0042] The method, device, equipment and medium for estimating the angle of arrival of an underwater acoustic array based on adaptive kernel width mixed correlation entropy proposed in the present application optimize the initial data matrix based on a preset mixed correlation entropy noise filtering method and an adaptive kernel width value to obtain a target data matrix; perform array angle of arrival estimation on the target data matrix to obtain spatial spectrum information for characterizing the target orientation; when the estimated angle difference between any signal sources in the spatial spectrum information is greater than the preset angle difference, the spatial spectrum information is determined to be a valid estimate, which solves the technical problem that the traditional underwater acoustic array DOA estimation method does not consider the non-Gaussian characteristics of the underwater environment, resulting in a decrease in estimation accuracy. Compared with the prior art, the improved noise filtering method based on mixed correlation entropy in the present application is applied to a multiple signal classification method, and the initial data matrix is ​​subspace decomposed by maximizing the mixed phase correlation entropy of the residual fitting error matrix, thereby effectively filtering out impulse noise. In an underwater environment dominated by non-Gaussian impulse noise, the estimation accuracy can also be effectively improved, thereby improving the detection capability of the underwater array. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings herein are incorporated in and constitute a part of the specification, illustrate embodiments consistent with the present application, and together with the description, are used to explain the principles of the present application.

[0044] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0045] Figure 1 A flowchart diagram of Embodiment 1 of the underwater acoustic array arrival angle estimation method based on adaptive kernel-width mixed correlation entropy provided in the present application;

[0046] Figure 2 A flow chart of Embodiment 2 of the method for estimating angle of arrival of an underwater acoustic array based on adaptive kernel-width mixed correlation entropy provided in the present application;

[0047] Figure 3 A schematic diagram of a specific process of an underwater acoustic array arrival angle estimation method based on adaptive kernel-width mixed correlation entropy provided in Example 2 of the present application;

[0048] Figure 4 RMSE curve of DOA estimation of the AMCC-MUSIC method under different kernel matrix width values ​​of the underwater acoustic array arrival angle estimation method based on adaptive kernel width mixed correlation entropy provided in Example 2 of the present application;

[0049] Figure 5 The resolution probability distribution of different algorithms of the underwater acoustic array arrival angle estimation method based on adaptive kernel width mixed correlation entropy provided in Example 2 of the present application;

[0050] Figure 6 RMSE curves of various algorithms under different GSNRs of the underwater acoustic array arrival angle estimation method based on adaptive kernel width mixed correlation entropy provided in Example 2 of the present application;

[0051] Figure 7 This is a schematic diagram of the module structure of an underwater acoustic array arrival angle estimation device based on adaptive kernel-width mixed correlation entropy according to an embodiment of the present application;

[0052] Figure 8 Schematic diagram of the device structure of the hardware operating environment involved in the underwater acoustic array arrival angle estimation method based on adaptive kernel-width mixed correlation entropy in the embodiment of the present application.

[0053] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0054] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0055] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0056] The main solution of the embodiment of the present application is: based on a preset hybrid correlation entropy noise filtering method and an adaptive kernel width value, an initial data matrix is ​​optimized to obtain a target data matrix; array arrival angle estimation is performed on the target data matrix to obtain spatial spectrum information for characterizing the target orientation; when the estimated angle difference between any signal sources in the spatial spectrum information is greater than the preset angle difference, the spatial spectrum information is determined to be a valid estimate.

[0057] It can be seen from the above embodiments that the present application optimizes the initial data matrix based on a preset hybrid correlation entropy noise filtering method and an adaptive kernel width value to obtain a target data matrix; the array arrival angle of the target data matrix is ​​estimated to obtain spatial spectrum information used to characterize the target orientation; when the estimated angle difference between any signal sources in the spatial spectrum information is greater than the preset angle difference, the spatial spectrum information is determined to be a valid estimate, which solves the technical problem that the traditional underwater acoustic array DOA estimation method does not consider the non-Gaussian characteristics of the underwater environment, resulting in a decrease in estimation accuracy. Compared with the prior art, the improved noise filtering method based on hybrid correlation entropy in the present application is applied to a multiple signal classification method, and the initial data matrix is ​​subspace decomposed by maximizing the hybrid phase correlation entropy of the residual fitting error matrix, thereby effectively filtering out the impulse noise. In an underwater environment dominated by non-Gaussian impulse noise, the estimation accuracy can also be effectively improved, thereby improving the detection capability of the underwater array.

[0058] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, an underwater acoustic array arrival angle estimation device based on adaptive kernel width mixed correlation entropy, etc. The following takes the underwater acoustic array arrival angle estimation based on adaptive kernel width mixed correlation entropy as an example to illustrate this embodiment and the following embodiments.

[0059] Based on this, the embodiment of the present application provides a method for estimating the angle of arrival of an underwater acoustic array based on adaptive kernel width mixed correlation entropy, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the underwater acoustic array arrival angle estimation method based on adaptive kernel-width mixed correlation entropy in the present application.

[0060] In this embodiment, the underwater acoustic array arrival angle estimation method based on adaptive kernel width mixed correlation entropy includes steps S10 to S20:

[0061] Step S10, based on a preset hybrid correlation entropy noise filtering method and an adaptive kernel width value, the initial data matrix is ​​optimized to obtain a target data matrix;

[0062] It should be noted that since the residual fitting error gradually decreases as the signal subspace is updated toward the steady state, a fixed kernel width may affect the adaptive dynamics and strike a balance between fast initial learning and fast convergence near the optimal point. Therefore, adaptively updating the kernel width based on the residual fitting error to obtain an adaptive kernel width can potentially improve the overall convergence speed and produce better steady-state performance.

[0063] In the specific implementation, 1. G and Z are two components of the received data matrix (initial data matrix), initializing G (i.e., the first random matrix) and Z matrix (i.e., the second random matrix), the number of iterations, and the iteration error ; 2. Check whether the iteration parameters (i.e., the number of iterations or the iteration error) meet the convergence conditions. If they do, exit the loop and input the target data matrix , if not satisfied, continue the loop; 3. Calculate high-order A matrix and its corresponding row vector ; 4. Calculate high-order A matrix and its corresponding column vector ; 5. Update high-order iteration parameter values , and repeat this cycle; 6. The two matrices that meet the convergence conditions indicate that the non-Gaussian pulse noise has been filtered out, and the received data matrix is ​​calculated and input into the MUSIC module; 7. The DOA of the array is estimated; 8. The spatial spectrum information representing the target direction is output.

[0064] Step S20: performing array arrival angle estimation on the target data matrix to obtain spatial spectrum information for characterizing the target orientation.

[0065] In a feasible implementation manner, the step of performing array arrival angle estimation on the target data matrix to obtain spatial spectrum information for characterizing the target position includes: determining a covariance matrix of the target data matrix; performing eigenvalue decomposition on the covariance matrix to obtain an eigenvector matrix; determining a signal subspace based on the eigenvector matrix, and determining a noise subspace based on the signal subspace; and based on the noise subspace, performing array arrival angle estimation on the target data matrix to obtain spatial spectrum information for characterizing the target position.

[0066] In the specific implementation, 1. Obtain the data optimization matrix after filtering out non-Gaussian noise (i.e. target data matrix):

[0067]

[0068] 2. Determine the covariance matrix of the target data matrix :

[0069]

[0070] 3. Covariance matrix Perform eigenvalue decomposition to obtain the eigenvector matrix :

[0071]

[0072] 4. Sort the eigenvalues ​​by size, the largest The eigenvalues ​​(assuming the number of sources is known) and the corresponding eigenvectors constitute the signal subspace, while the remaining eigenvectors constitute the noise subspace. So based on the MUSIC method, DOA is estimated by searching for spatial spectrum peaks:

[0073]

[0074] In the formula, (·) H represents the conjugate transpose, and a(θ) represents the array steering vector.

[0075] Step S30: When the estimated angle differences between any signal sources in the spatial spectrum information are greater than a preset angle difference, the spatial spectrum information is determined to be a valid estimate.

[0076] It should be noted that, in order to improve the DOA estimation accuracy, a preset angle difference can be set in advance. When the estimated angle difference between any two signal sources in the spatial spectrum information is less than or equal to the preset angle difference, the spatial spectrum information output this time is determined to be an invalid estimate and needs to be recalculated. When the estimated angle difference between any two signal sources in the spatial spectrum information is greater than the preset angle difference, the spatial spectrum information output this time can be determined to be a valid estimate. If the estimated angle between any two signal sources in the spatial spectrum information is less than or equal to the preset angle difference (if ), the spatial spectrum information is determined to be an invalid estimate, and the algorithm recalculates the spatial spectrum information.

[0077] This embodiment optimizes the initial data matrix based on a preset hybrid correlation entropy noise filtering method and an adaptive kernel width value to obtain a target data matrix; performs array arrival angle estimation on the target data matrix to obtain spatial spectrum information for characterizing the target orientation; when the estimated angle difference between any signal sources in the spatial spectrum information is greater than the preset angle difference, the spatial spectrum information is determined to be a valid estimate, which solves the technical problem that the traditional underwater acoustic array DOA estimation method does not consider the non-Gaussian characteristics of the underwater environment, resulting in a decrease in estimation accuracy. Compared with the prior art, the improved noise filtering method based on hybrid correlation entropy in the present application is applied to a multiple signal classification method, and the initial data matrix is ​​subspace decomposed by maximizing the hybrid correlation entropy of the residual fitting error matrix, thereby effectively filtering out impulse noise. In an underwater environment dominated by non-Gaussian impulse noise, the estimation accuracy can also be effectively improved, thereby improving the detection capability of the underwater array.

[0078] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can be referred to the above introduction, and will not be repeated in the following. Figure 2 , step S10 also includes steps S101 to S104:

[0079] S101, receiving input data, wherein the input data includes an initial data matrix, a maximum number of iterations, and a maximum allowable relative error;

[0080] It should be noted that the input data includes the initial data matrix , the maximum number of iterations T1 and the maximum allowable relative error .

[0081] S102, initializing a first random matrix, a second random matrix and an iteration parameter according to the initial data matrix, wherein the first random matrix is ​​a full column rank random matrix, and the second random matrix is ​​a full row rank random matrix;

[0082] It should be noted that the iteration parameters include the number of iterations and the iteration error, and initialize a full-column rank random matrix , a full row rank random matrix .

[0083] S103, calculating a first high-order matrix of the first random matrix and calculating a second high-order matrix of the second random matrix based on a preset hybrid correlation entropy noise filtering method and an adaptive kernel width value;

[0084] It should be noted that, assuming there are K far-field sound sources s k(n), where k = 1, 2, ..., K, produces line spectrum signals of the same frequency. These signals are received in the far field by a passive sonar ULA consisting of M omnidirectional array sensors. The incident directions of the K far-field sound sources are θ1, θ2, ..., θ k The spacing between the array sensors is equal to half the signal wavelength. At the nth snapshot moment, the data vector x(n) received by the array can be expressed as:

[0085]

[0086] Among them, x(n) = [x1(n), x2(n),..., x M (n)] T is the data vector of the nth snapshot, s(n) = [s1(n), ..., s K (n)] T ,s(n) = [s1(n), ..., s K (n)] T N is the maximum number of snapshots of the signal. n(n) = [n1(n),n2(n), ..., n K (n)] T represents additive background noise. The matrix A(θ) = [a(θ1),..., a(θ K )] is the steering vector matrix of the array. The steering vector for the kth source (which describes the directional response) is:

[0087]

[0088] Therefore, the data matrix of the N snapshots received by the array can be expressed as:

[0089]

[0090] Among them, A∈C M×K represents the steering matrix of the array, S = [s(1), ..., s(N)]∈C K×N represents the incident signal matrix, N = [n(1), ..., n(N)]∈C M×N represents the additive noise matrix received by the array.

[0091] It should be noted that in the field of DOA estimation, especially in methods involving subspace decomposition, it is usually required that the number of elements M in the array exceeds the number of signal sources received by the array, that is, M>K. Therefore, the matrix A is full column rank and the matrix S is full row rank, and the rank of both is K. Assuming that the noise matrix N received by the array is a zero matrix, the data matrix X received by the array can be decomposed into:

[0092]

[0093] Among them, the matrix G∈C M×K The rank is K, the matrix Z∈C K×N The rank of is also K. In addition, the column vectors of matrix G and the column vectors of matrix A belong to the same vector space.

[0094] When the noise received by the array follows a Gaussian distribution, it is recommended to consider minimizing the squared error function:

[0095]

[0096] Among them, X mn and (GZ) mn denote the (m,n)th entry of matrices X and GZ, respectively. Here, e mn =X mn −(GZ) mn represents the residual fitting error, X−GZ is the residual matrix, and ||.|| F Denotes the Frobenius norm.

[0097] In a feasible implementation, the steps of calculating the first high-order matrix of the first random matrix based on a preset mixed correlation entropy noise filtering method and an adaptive kernel width value, and calculating the second high-order matrix of the second random matrix include: based on the preset mixed correlation entropy noise filtering method, constructing a cost function of the maximum mixed correlation entropy criterion according to a Gaussian function and an adaptive kernel width value; based on the cost function, calculating the first high-order matrix of the first random matrix and calculating the second high-order matrix of the second random matrix through an alternating optimization strategy.

[0098] It should be noted that in order to design a more robust DOA estimator based on the subspace method in a non-Gaussian pulse noise environment, it is feasible to use MMCC to replace the Frobenius norm in formula (4) (that is, the pulse noise filtering method based on mixed entropy uses mixed entropy to represent the cost function in the MMCC method). The new cost function expression is as follows (cost function):

[0099]

[0100] in , represents the jth order mixing coefficient, satisfying r1+r2=1. Among them, the Gaussian function k δj (e mn ) is calculated as:

[0101]

[0102] Among them, δj is a Gaussian function, k δj (e mn ) is the kernel width. According to MMCC, maximizing formula (5) helps to estimate G and Z.

[0103] It should be noted that since the kernel width is a key free parameter in MMCC adaptive filtering and has an important influence on the convergence speed and steady-state performance, the kernel width of the adaptive update is set, and the square of the kernel width is set to the sum of the square of the preset kernel width and the square of the residual fitting error. This new method achieves better adaptive filtering performance while maintaining the simplicity and robustness of the MMCC algorithm.

[0104] Combined with formula (5), the cost function for updating the (m,n)th element in matrices G and Z can be concisely expressed as:

[0105]

[0106] Where g and z represent the (m,n)th element of matrices G and Z respectively, and e mn Abbreviated as e.

[0107] According to MMCC, find the maximum value of A C The kernel width of (g, z) is equivalent to the following optimization problem:

[0108]

[0109] Obviously, when:

[0110]

[0111] At this time, the corresponding adaptive core width meets the requirements of MMCC.

[0112] For function A C Taking the derivative of (g,z) with respect to the kernel width δ1, we get:

[0113]

[0114] Therefore, when:

[0115]

[0116] In a feasible implementation, the step of constructing a cost function of the maximum mixed related entropy criterion based on a Gaussian function and an adaptive kernel width value includes: determining a residual fitting error value of a current iteration; determining an adaptive kernel width value based on the residual fitting error value, a proportional coefficient, and a preset kernel width; and constructing a cost function of the maximum mixed related entropy criterion based on the Gaussian function and the adaptive kernel width value.

[0117] It can be understood that in order to adapt to the specific situation of impulse noise, the adaptive rule of mixed entropy is modified, and the adaptive kernel width of the mixed entropy value corresponding to 0 in the above formula is expressed as:

[0118]

[0119] In the formula, and represents the preset kernel width, It represents the proportionality coefficient, which can be set according to the simulation experiment results. Preferably, the proportionality coefficient can be set to 0.5.

[0120] It should be noted that in the framework of mixed correlation entropy (MMCC), how to update the iterative result G through alternating optimization strategies (t+1) and Z (t+1) :

[0121] According to the alternating optimization strategy in MMCC, the next iteration result G can be derived by solving the following problem: (t +1) and Z (t+1) :

[0122]

[0123]

[0124] Combining formulas (5), (13), and (14), it can be specifically expressed as follows:

[0125]

[0126]

[0127] in, and Respectively represent matrices and No. List, and Respectively represent matrices and No. OK.

[0128] In a feasible implementation, the steps of calculating the first high-order matrix of the first random matrix through an alternating optimization strategy based on the cost function, and calculating the second high-order matrix of the second random matrix include: iteratively determining the first independent sub-item according to the first random matrix based on the cost function and the alternating optimization strategy, and determining the first high-order matrix according to the first independent sub-item; iteratively determining the second independent sub-item according to the second random matrix based on the cost function and the alternating optimization strategy, and determining the first high-order matrix according to the second independent sub-item.

[0129] It should be noted that in formula (16), the row vectors of matrices X and Z are independent, so formula (16) can be decomposed into M independent sub-problems, and each sub-problem can be solved separately as follows:

[0130]

[0131] definition , , , , for the function Taking the derivative, we get:

[0132]

[0133] in, is a diagonal matrix with the first The elements are represented as:

[0134]

[0135] Solved by stochastic gradient method The maximum value of It can be calculated iteratively by the following formula:

[0136]

[0137] in is the step length, which can be calculated by the following formula:

[0138]

[0139] Among them, ||.||2 represents the l2 norm.

[0140] S104. Update the iteration parameters according to the first random matrix, the first high-order matrix, the second random matrix and the second high-order matrix, and use the first high-order matrix as a new first random matrix and the second high-order matrix as a new second random matrix to repeat steps S103-S104 until the current iteration parameters meet the convergence conditions, use the current first high-order matrix and the current second high-order matrix as the first optimal matrix and the second optimal matrix respectively, and determine the target data matrix according to the first optimal matrix and the second optimal matrix, wherein the iteration parameters include the number of iterations and the iteration error, and the convergence condition is that the current number of iterations is greater than the maximum number of iterations or the current iteration error is greater than the maximum allowable relative error.

[0141] In the specific implementation, 1. and are the two components of the received data matrix (initial data matrix), initialized (i.e. the first random matrix) and Matrix (i.e., the second random matrix), number of iterations, and iteration error ; 2. Check whether the iteration parameters (i.e., the number of iterations or the iteration error) meet the convergence conditions. If they do, exit the loop and input the target data matrix , if not satisfied, continue the loop; 3. Calculate high-order A matrix and its corresponding row vector ; 4. Calculate high-order A matrix and its corresponding column vector ; 5. Update high-order iteration parameter values , and repeat this cycle; 6. The two matrices that meet the convergence conditions indicate that the non-Gaussian pulse noise has been filtered out, and the received data matrix is ​​calculated and input into the MUSIC module; 7. The DOA of the array is estimated; 8. The spatial spectrum information representing the target direction is output.

[0142] In the specific implementation, Figure 3 As shown, receiving input data, wherein the input data includes a data matrix , maximum number of iterations , Maximum allowable relative error ;

[0143] Initialize a full column rank random matrix , a full row rank random matrix

[0144] 1:for do

[0145] 2:Step 1: Iterative Optimization

[0146]

[0147] 3:Step 2: Solve the independent sub-items

[0148]

[0149] 4:Step 3: Iteratively calculate the second optimization item

[0150]

[0151] 5:Step 4: Iterative solution Second-rate

[0152]

[0153] 6:Step 5:Update And calculate the relative error

[0154]

[0155] 7: end for

[0156] 8:Step 6: Repeat Step 1 to Step 5 until or

[0157] 9: Output: Final optimization matrix and (i.e. the first optimal matrix and the second optimal matrix)

[0158] In a feasible implementation, the step of updating the iteration parameters according to the first random matrix, the first high-order matrix, the second random matrix and the second high-order matrix includes: determining a first cost function value based on the cost function, the first random matrix and the second random matrix; determining a second cost function value based on the cost function, the first high-order matrix and the second high-order matrix; determining a relative error according to the first cost function value and the second cost function value; updating the iteration error according to the relative error, and updating the number of iterations.

[0159] In the specific implementation, the iterative error (relative error) is calculated as follows:

[0160]

[0161] in, represents the iteration error, represents the first cost function value, Represents the second cost function value, and the number of iterations is updated as follows: t=t+1.

[0162] In the specific implementation, a concept of measuring the relative strength of the signal and the impulse noise is introduced in the performance analysis, and its expression is:

[0163]

[0164] in, Represents the dispersion coefficient.

[0165] In the specific implementation, the root mean square error (RMSE) is used to quantitatively represent the DOA estimation accuracy, and its expression is:

[0166]

[0167] It should be noted that if Figure 4 The RMSE curves of DOA estimation of the AMCC-MUSIC method under different kernel matrix width values ​​are shown. The results show that the DOA estimation performance of AMCC-MUSIC is not sensitive to the initial setting of the kernel matrix width. In subsequent simulations, the initial kernel matrix width value is uniformly set. .

[0168] In the specific implementation, the resolution probability is defined as the number of experiments that successfully estimate the target DOA in independent experiments, expressed as:

[0169]

[0170] Where T represents the number of independent experiments, T S Represents the number of experiments that successfully estimated the DOA.

[0171] It should be noted that if Figure 5 The resolution probability distribution of different algorithms is shown in Figure 5 Two center frequencies are set in the simulation. The incoherent narrowband Gaussian white noise signal source is The direction of the incident light is incident on the array. From the simulation results, it can be seen that the AMCC-MUSIC algorithm proposed in this scheme has a higher resolution performance than the traditional algorithm.

[0172] It should be noted that if Figure 6The figure shows the RMSE curves of each algorithm under different GSNR. From the simulation results, it can be seen that the traditional MCC-MUSIC algorithm with a fixed kernel matrix width does not converge as the GSNR increases, while the MMCC-MUSIC algorithm has a better convergence. Therefore, the MUSIC algorithm using AMCC has better robustness than the MCC-MUSIC algorithm with a fixed kernel matrix width. The flexibility of the AMCC-MUSIC algorithm in kernel width selection relative to the MCC-MUSIC algorithm determines the superiority of the entropy signal in signal processing performance.

[0173] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the underwater acoustic array arrival angle estimation method based on adaptive kernel-width mixed correlation entropy in the present application. More forms of simple transformations based on this technical concept are all within the protection scope of the present application.

[0174] The present application also provides an underwater acoustic array arrival angle estimation device based on adaptive kernel width mixed correlation entropy, please refer to Figure 7 , the underwater acoustic array arrival angle estimation device based on adaptive kernel width mixed correlation entropy includes:

[0175] An optimization module 10 is used to optimize the initial data matrix based on a preset mixed correlation entropy noise filtering method and an adaptive kernel width value to obtain a target data matrix;

[0176] The output module 20 is used to estimate the array arrival angle of the target data matrix to obtain spatial spectrum information for characterizing the target orientation.

[0177] The determination module 30 is configured to determine that the spatial spectrum information is a valid estimate when the estimated angle differences between any signal sources in the spatial spectrum information are greater than a preset angle difference.

[0178] The underwater acoustic array arrival angle estimation device based on adaptive kernel-width mixed correlation entropy provided in the present application adopts the underwater acoustic array arrival angle estimation method based on adaptive kernel-width mixed correlation entropy in the above-mentioned embodiment, which can solve the technical problem that the traditional underwater acoustic array DOA estimation method does not consider the non-Gaussian characteristics of the underwater environment, resulting in a decrease in estimation accuracy. Compared with the prior art, the beneficial effects of the underwater acoustic array arrival angle estimation device based on adaptive kernel-width mixed correlation entropy provided in the present application are the same as the beneficial effects of the underwater acoustic array arrival angle estimation method based on adaptive kernel-width mixed correlation entropy provided in the above-mentioned embodiment, and the other technical features of the underwater acoustic array arrival angle estimation device based on adaptive kernel-width mixed correlation entropy are the same as the features disclosed in the above-mentioned embodiment method, which will not be repeated here.

[0179] The present application provides an underwater acoustic array arrival angle estimation device based on adaptive kernel-width mixed correlation entropy. The underwater acoustic array arrival angle estimation device based on adaptive kernel-width mixed correlation entropy includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the underwater acoustic array arrival angle estimation method based on adaptive kernel-width mixed correlation entropy in the above-mentioned embodiment one.

[0180] Reference below Figure 8 , which shows a schematic diagram of the structure of an underwater acoustic array arrival angle estimation device based on adaptive kernel-width mixed correlation entropy suitable for implementing the embodiment of the present application. The underwater acoustic array arrival angle estimation device based on adaptive kernel-width mixed correlation entropy in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions: tablet computers), PMPs (Portable Media Players: portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 8 The underwater acoustic array arrival angle estimation device based on adaptive kernel width mixed correlation entropy shown is only an example and should not bring any limitation to the function and scope of use of the embodiments of the present application.

[0181] like Figure 8As shown, the underwater acoustic array arrival angle estimation device based on adaptive kernel width mixed correlation entropy may include a processing device 1001 (such as a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM: ReadOnly Memory) 1002 or the program loaded from the storage device 1003 to the random access memory (RAM: RandomAccess Memory) 1004. Various programs and data required for the operation of the underwater acoustic array arrival angle estimation device based on adaptive kernel width mixed correlation entropy are also stored in RAM1004. The processing device 1001, ROM1002 and RAM1004 are connected to each other through a bus 1005. The input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 can allow the water acoustic array arrival angle estimation device based on adaptive kernel width mixed correlation entropy to communicate wirelessly or wired with other devices to exchange data. Although the figure shows a water acoustic array arrival angle estimation device based on adaptive kernel width mixed correlation entropy with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have instead.

[0182] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0183] The underwater acoustic array arrival angle estimation device based on adaptive kernel-width mixed correlation entropy provided by the present application adopts the underwater acoustic array arrival angle estimation method based on adaptive kernel-width mixed correlation entropy in the above-mentioned embodiment, which can solve the technical problem that the traditional underwater acoustic array DOA estimation method does not consider the non-Gaussian characteristics of the underwater environment, resulting in a decrease in estimation accuracy. Compared with the prior art, the beneficial effects of the underwater acoustic array arrival angle estimation device based on adaptive kernel-width mixed correlation entropy provided by the present application are the same as the beneficial effects of the underwater acoustic array arrival angle estimation method based on adaptive kernel-width mixed correlation entropy provided by the above-mentioned embodiment, and the other technical features of the underwater acoustic array arrival angle estimation device based on adaptive kernel-width mixed correlation entropy are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0184] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0185] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

[0186] The present application provides a computer-readable storage medium having computer-readable program instructions (i.e., computer programs) stored thereon, and the computer-readable program instructions are used to execute the underwater acoustic array arrival angle estimation method based on adaptive kernel-width mixed correlation entropy in the above-mentioned embodiment.

[0187] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM: Random Access Memory), a read-only memory (ROM: Read Only Memory), an erasable programmable read-only memory (EPROM: Erasable Programmable Read Only Memory or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM: CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency: Radio Frequency), etc., or any suitable combination of the above.

[0188] The above-mentioned computer-readable storage medium can be included in the underwater acoustic array arrival angle estimation device based on adaptive kernel-width mixed correlation entropy; or it can exist independently without being assembled into the underwater acoustic array arrival angle estimation device based on adaptive kernel-width mixed correlation entropy.

[0189] The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by the underwater acoustic array arrival angle estimation device based on adaptive kernel width mixed correlation entropy, the underwater acoustic array arrival angle estimation device based on adaptive kernel width mixed correlation entropy: optimizes the initial data matrix based on the preset mixed correlation entropy noise filtering method and the adaptive kernel width value to obtain the target data matrix; performs array arrival angle estimation on the target data matrix based on the spatial spectrum peak search of the multiple signal classification method, and obtains the spatial spectrum information used to characterize the target position.

[0190] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0191] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0192] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0193] The readable storage medium provided in the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned method for estimating the angle of arrival of an underwater acoustic array based on adaptive kernel-width mixed correlation entropy, and can solve the technical problem that the traditional method for estimating the DOA of an underwater acoustic array does not consider the non-Gaussian characteristics of the underwater environment, resulting in a decrease in estimation accuracy. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in the present application are the same as the beneficial effects of the method for estimating the angle of arrival of an underwater acoustic array based on adaptive kernel-width mixed correlation entropy provided in the above-mentioned embodiment, and will not be described in detail here.

[0194] The present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the above-mentioned method for estimating the angle of arrival of an underwater acoustic array based on adaptive kernel-width mixed correlation entropy.

[0195] The computer program product provided by the present application can solve the technical problem that the traditional underwater acoustic array DOA estimation method does not consider the non-Gaussian characteristics of the underwater environment, resulting in a decrease in estimation accuracy. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as the beneficial effects of the underwater acoustic array arrival angle estimation method based on adaptive kernel width mixed correlation entropy provided in the above embodiment, and will not be repeated here.

[0196] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A method for estimating angle of arrival of underwater acoustic array based on adaptive kernel width mixed correlation entropy, characterized in that: The method comprises: Based on a preset mixed correlation entropy noise filtering method and an adaptive kernel width value, an initial data matrix is ​​optimized to obtain a target data matrix, wherein the target data matrix is ​​obtained by optimizing the initial data matrix through an alternating optimization strategy based on a cost function, the cost function is a cost function of a maximum mixed correlation entropy criterion constructed according to a Gaussian function and an adaptive kernel width value based on the preset mixed correlation entropy noise filtering method, and the adaptive kernel width value is determined based on a residual fitting error value, a proportional coefficient, and a preset kernel width in an iterative optimization process; Performing array arrival angle estimation on the target data matrix to obtain spatial spectrum information for characterizing the target orientation; When the estimated angle differences between any signal sources in the spatial spectrum information are greater than a preset angle difference, the spatial spectrum information is determined to be a valid estimate.

2. The method according to claim 1, characterized in that The step of optimizing the initial data matrix based on the preset mixed correlation entropy noise filtering method and the adaptive kernel width value to obtain the target data matrix includes: S101, receiving input data, wherein the input data includes an initial data matrix, a maximum number of iterations, and a maximum allowable relative error; S102, initializing a first random matrix, a second random matrix and an iteration parameter according to the initial data matrix, wherein the first random matrix is ​​a full column rank random matrix, and the second random matrix is ​​a full row rank random matrix; S103, calculating a first high-order matrix of the first random matrix and calculating a second high-order matrix of the second random matrix based on a preset hybrid correlation entropy noise filtering method and an adaptive kernel width value; S104. Update the iteration parameters according to the first random matrix, the first high-order matrix, the second random matrix and the second high-order matrix, and use the first high-order matrix as a new first random matrix and the second high-order matrix as a new second random matrix to repeat steps S103-S104 until the current iteration parameters meet the convergence conditions, use the current first high-order matrix and the current second high-order matrix as the first optimal matrix and the second optimal matrix respectively, and determine the target data matrix according to the first optimal matrix and the second optimal matrix, wherein the iteration parameters include the number of iterations and the iteration error, and the convergence condition is that the current number of iterations is greater than the maximum number of iterations or the current iteration error is greater than the maximum allowable relative error.

3. The method according to claim 2, characterized in that The steps of calculating the first high-order matrix of the first random matrix and calculating the second high-order matrix of the second random matrix based on the preset hybrid correlation entropy noise filtering method and the adaptive kernel width value include: Based on the preset mixed correlation entropy noise filtering method, a cost function of a maximum mixed correlation entropy criterion is constructed according to a Gaussian function and an adaptive kernel width value; Based on the cost function, a first high-order matrix of the first random matrix is ​​calculated through an alternating optimization strategy, and a second high-order matrix of the second random matrix is ​​calculated.

4. The method according to claim 3, characterized in that The step of constructing a cost function of the maximum mixed correlation entropy criterion according to the Gaussian function and the adaptive kernel width value comprises: Determine the residual fitting error value of the current iteration; Determining an adaptive kernel width value based on the residual fitting error value, the proportionality coefficient and the preset kernel width; A cost function of a maximum mixed correlation entropy criterion is constructed according to the Gaussian function and the adaptive kernel width value.

5. The method according to claim 3, characterized in that The steps of calculating the first high-order matrix of the first random matrix and calculating the second high-order matrix of the second random matrix by an alternating optimization strategy based on the cost function include: Based on the cost function and the alternating optimization strategy, iteratively determine a first independent sub-item according to the first random matrix, and determine a first high-order matrix according to the first independent sub-item; Based on the cost function and the alternating optimization strategy, a second independent sub-item is iteratively determined according to the second random matrix, and a first high-order matrix is ​​determined according to the second independent sub-item.

6. The method according to claim 3, characterized in that The step of updating the iteration parameter according to the first random matrix, the first high-order matrix, the second random matrix and the second high-order matrix comprises: Determining a first cost function value based on the cost function, the first random matrix, and the second random matrix; Determining a second cost function value based on the cost function, the first high-order matrix, and the second high-order matrix; Determining a relative error according to the first cost function value and the second cost function value; The iteration error is updated according to the relative error, and the number of iterations is updated.

7. The method according to claim 1, characterized in that The step of performing array arrival angle estimation on the target data matrix to obtain spatial spectrum information for characterizing the target orientation comprises: Determining a covariance matrix of the target data matrix; Performing eigenvalue decomposition on the covariance matrix to obtain an eigenvector matrix; Determine a signal subspace according to the eigenvector matrix, and determine a noise subspace according to the signal subspace; Based on the noise subspace, the array arrival angle of the target data matrix is ​​estimated to obtain spatial spectrum information for characterizing the target orientation.

8. An underwater acoustic array arrival angle estimation device based on adaptive kernel width mixed correlation entropy, characterized in that: The device comprises: An optimization module is used to optimize the initial data matrix based on a preset mixed correlation entropy noise filtering method and an adaptive kernel width value to obtain a target data matrix, wherein the target data matrix is ​​obtained by optimizing the initial data matrix through an alternating optimization strategy based on a cost function, the cost function is a cost function of a maximum mixed correlation entropy criterion constructed according to a Gaussian function and an adaptive kernel width value based on the preset mixed correlation entropy noise filtering method, and the adaptive kernel width value is determined based on a residual fitting error value, a proportional coefficient, and a preset kernel width in an iterative optimization process; An output module is used to estimate the array arrival angle of the target data matrix to obtain spatial spectrum information for characterizing the target orientation; The determination module is configured to determine that the spatial spectrum information is a valid estimation when the estimated angle differences between any signal sources in the spatial spectrum information are greater than a preset angle difference.

9. An underwater acoustic array arrival angle estimation device based on adaptive kernel width mixed correlation entropy, characterized in that: The device comprises: a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the underwater acoustic array arrival angle estimation method based on adaptive kernel-width mixed correlation entropy as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the underwater acoustic array arrival angle estimation method based on adaptive kernel-width mixed correlation entropy as described in any one of claims 1 to 7 are implemented.

Images