Vector hydrophone motion array doa estimation method

Abstract

The present application belongs to the field of underwater target direction estimation, and relates to a vector hydrophone moving array DOA estimation method with array element angle deviation, comprising the following steps: 1) obtaining an array element angle deviation matrix and performing coarse estimation on the array element angle deviation matrix; 2) reconstructing the coarse estimation result of the array element angle deviation matrix to obtain a reconstructed angle deviation matrix; 3) calculating an estimated value of a signal matrix based on the angle deviation matrix; alternately iterating the estimated value of the signal matrix and the reconstructed angle deviation matrix to obtain a restored array receiving signal matrix; and 4) performing vector hydrophone array ETAM algorithm and beam forming based on the restored array receiving signal matrix to estimate the target direction. The present application provides a vector hydrophone moving array DOA estimation method which can effectively improve the estimation accuracy.

Description

Technical Field

[0001] This invention belongs to the field of underwater target orientation estimation, and relates to a method for estimating the DOA of a moving vector hydrophone array, and more particularly to a method for estimating the DOA of a moving vector hydrophone array with element angle deviation. Background Technology

[0002] Ocean development is crucial to humankind. Since seawater provides the best sound propagation medium, advancements in underwater acoustic technology are indispensable for ocean exploration. By analyzing the acoustic information received by hydrophone arrays, the location of underwater targets can be estimated. Unmanned and intelligent technologies have become important trends in modern warfare. Unmanned Underwater Vehicles (UUVs) are essential equipment for detecting marine resources, conducting underwater intelligence gathering, and safeguarding maritime security, playing an irreplaceable role in estimating the location of underwater targets.

[0003] A UUV (Underwater Vehicle) is a small, easily deployed and recovered underwater vehicle capable of autonomous, long-range underwater navigation. UUVs have a wide range of operational applications, high autonomy, and strong maneuverability, enabling them to perform underwater search and measurement, underwater battlefield prediction, battlefield surveillance and analysis, and underwater auxiliary communication. Due to the continuous reduction in radiated noise from underwater targets in recent years, improving the high-resolution detection capability of UUVs against underwater targets is one of the key technological challenges that countries worldwide are vying to solve.

[0004] Synthetic aperture technology (SAR) in China can significantly improve the resolution of underwater target detection. Traditional sonar systems are limited in resolution due to beamwidth and signal propagation distance, while SAR radar, by transmitting wideband signals and utilizing the received echo signals, combined with information about the radar platform's movement, can reconstruct high-resolution images. SAR technology can be used in small arrays, achieving resolution and gain several times or even tens of times greater than physical arrays. Furthermore, vector hydrophones also offer excellent spatial gain and resolution. Therefore, the fusion of vector hydrophone array technology and SAR technology can effectively improve target location estimation accuracy and resolution. Research on SAR sonar technology in China began in the late 1990s. In 2000, Sun Dajun et al. reviewed the current research status of SAR technology in underwater acoustic applications, focusing on key issues such as media stability, motion compensation, mapping rate, and imaging algorithms. From 2008 to 2010, the ETAM algorithm made significant progress. In 2010, Hou Yunshan et al. proposed a synthetic aperture algorithm based on a robust Capon beamformer, improving the anti-interference capability of the ETAM algorithm. In 2014, Yu Tongkui et al. applied synthetic aperture technology to vector arrays, demonstrating that the vector array synthetic aperture algorithm has higher target positioning accuracy, array gain, and azimuth resolution, making it suitable for measuring radiated noise from low-noise targets. In 2016, Zhang Zhexian et al. proposed a synthetic aperture circular array algorithm based on differential evolution, providing valuable reference for optimizing element arrangement when synthesizing apertures of unequally spaced arrays. In 2018, Tian Yingze studied a vector array detection method based on a small platform, analyzing the performance of the synthetic aperture algorithm and virtual element technology, and highlighting the impact of non-ideal conditions on the performance of the synthetic aperture algorithm.

[0005] Because of the unpredictable movement of UUVs underwater, it is difficult to ensure that they move on the same horizontal plane during the synthesis of the aperture. This introduces a consistent angular deviation to each array element during the synthesis process. Furthermore, since vector hydrophones contain velocity sensors, it is difficult to ensure that the velocity sensors of each element point in the same direction on the X and Y axes when installing the vector hydrophone array, leading to random axial deviations for each element. These two factors, resulting in element angular deviations, degrade the detection performance of the vector hydrophone array based on the moving platform. Summary of the Invention

[0006] To address the aforementioned technical problems in the background art, this invention provides a vector hydrophone motion array DOA estimation method that can effectively improve estimation accuracy.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] A method for estimating the DOA of a moving array of a vector hydrophone, characterized in that the method includes the following steps:

[0009] 1) Obtain the element angle deviation matrix and perform a coarse estimation of the element angle deviation matrix;

[0010] 2) Reconstruct the coarse estimation result of the array element angle deviation matrix obtained in step 1) to obtain the reconstructed angle deviation matrix;

[0011] 3) Calculate the estimated value of the signal matrix from the angle deviation matrix obtained in step 2); iterate the estimated value of the signal matrix and the reconstructed angle deviation matrix obtained in step 2) alternately to obtain the restored array received signal matrix;

[0012] 4) Based on the restored array received signal matrix obtained in step 3), perform the vector hydrophone array ETAM algorithm and beamforming to estimate the target azimuth.

[0013] Preferably, in step 1) of the present invention, the array element angle deviation matrix is ​​G, and the form in which the array element angle deviation matrix G is expressed is:

[0014]

[0015] The result of the coarse estimation of the pair element angle deviation matrix is The The expression is:

[0016]

[0017] in:

[0018] g is the angular deviation matrix of the vector hydrophone;

[0019] N is the total number of array elements;

[0020] X is the data matrix received by the vector hydrophone array;

[0021] S is the array manifold matrix;

[0022] It is the target signal vector.

[0023] Preferably, the specific implementation of step 2) in this invention is as follows:

[0024] 2.1) Results of coarse estimation of the pair element angle deviation matrix

[0025] Normalization is performed to obtain the normalized result.

[0026] 2.2) Normalize the result obtained in step 2.1). Divide into H 2×2 sub-blocks, where the h-th sub-block is represented as... Where h≤H; the The expression is:

[0027]

[0028] in:

[0029] α = 3h-1;

[0030] and It is related to the cosine value of the array element angle deviation;

[0031] and It is related to the sine value of the array element angle deviation;

[0032] It is the α-th row and α-th column of the normalized result;

[0033] It represents the α+1th row and α+1th column of the normalized result;

[0034] yes The normalization result;

[0035] 2.3) After performing the corresponding inverse cosine and arcsine transformations on each element of the pairwise angle deviation matrix, rewrite the matrix, denoted as . Where i represents the i-th iteration, the The expression is:

[0036]

[0037] 2.4) Taking the first element in the array element angle deviation matrix as a reference, first solve for the relative deviation matrix of the nth vector hydrophone, denoted as […]. The The expression is:

[0038]

[0039] Where: Re is the real part of the complex number;

[0040] 2.5) The relative deviation matrix of the vector hydrophone obtained in step 2.4). Based on this, construct the angle deviation matrix of the vector hydrophone. The angular deviation matrix of the vector hydrophone The expression is:

[0041]

[0042] in:

[0043] It is the estimated angular deviation value of the nth vector hydrophone;

[0044] The The expression is:

[0045]

[0046] in:

[0047] as well as The angle deviation β is obtained by using the sine and cosine functions, respectively. n The estimated value;

[0048] The The expression is:

[0049]

[0050] The The expression is:

[0051]

[0052] 2.6) The angular deviation matrix of the vector hydrophone obtained in step 2.5). The angle deviation matrix of the basic pair is reconstructed to obtain the reconstructed angle deviation matrix. The reconstructed angle deviation matrix The expression is:

[0053]

[0054] in:

[0055] n≤N.

[0056] Preferably, the specific implementation of step 3) in this invention is as follows:

[0057] 3.1) Obtaining the signal The It is a signal matrix. The m-th vector in the middle;

[0058] 3.2) Based on the reconstructed angle deviation matrix obtained in step 2), The signal obtained in step 3.1) Make predictions and obtain signals The estimated value

[0059] 3.3) The signal estimate and the angle deviation matrix are iterated alternately.

[0060] Preferably, the signal in step 3.1) of the present invention The expression is:

[0061]

[0062] in:

[0063] X = [x(1), ..., x(L)] T X represents the data matrix received by the vector hydrophone array;

[0064] T stands for transpose;

[0065]

[0066] R is the covariance matrix of the data received by the acoustic bar hydrophone array;

[0067] It is an array manifold matrix;

[0068] H is the conjugate transpose, which is the conjugate transpose of A;

[0069] pm represents the m-th row in the signal power matrix P; the signal power matrix is ​​P = [p1, p2, ..., pm]. M ] T ;

[0070] Where m is 1, 2, 3, ..., M;

[0071] L represents the number of snapshots within the measurement time;

[0072]

[0073] i represents the i-th iteration.

[0074] Preferably, the signal in step 3.2) of the present invention The estimated value The expression is:

[0075]

[0076] in:

[0077] It is a weighted compensation factor; The expression is:

[0078]

[0079] in:

[0080] L represents the number of snapshots within the measurement time;

[0081] N represents the number of elements in the vector hydrophone array;

[0082] It is the estimated value of pm in the i-th iteration;

[0083] These are process derivation parameters, the stated The expression is:

[0084]

[0085] in:

[0086] X represents the data matrix received by the vector hydrophone array;

[0087] Tr represents the trace of the matrix;

[0088] These are process derivation parameters, the stated The expression is:

[0089]

[0090] Preferably, the signal power corresponding to the scanning area in step 3.3) of the present invention is The The expression is:

[0091]

[0092] in:

[0093] It is an iteration of the signal, and its expression is:

[0094] Preferably, the convergence condition for the iteration in step 3.4) of the present invention is: the difference between the angle deviation matrix and the identity matrix is ​​taken, and if the second norm of the difference matrix is ​​less than 0.001, the iteration loop is exited.

[0095] The advantages of this invention are:

[0096] This invention utilizes the proposed ETAM algorithm to improve the azimuth estimation performance of a moving platform vector hydrophone array with element angle deviations, thereby increasing the accuracy of azimuth estimation. The key to this method is leveraging the inherent directivity of the vector hydrophone to reconstruct the coarse estimation result, making the angle deviation matrix estimation result more accurate. This invention provides a DOA estimation method for moving vector hydrophone arrays with element angle deviations, combining a moving platform with a vector hydrophone synthetic aperture algorithm to improve the azimuth estimation accuracy of the ETAM algorithm under vector hydrophone arrays. The method includes the following steps: 1) First, a coarse estimation of the element angle deviation matrix is ​​performed, and then the inherent directivity of the vector hydrophone is used to reconstruct the coarse estimation result, making the angle deviation matrix estimation result more accurate. 2) The angle deviation matrix is ​​fixed, and the signal is recovered. After alternating iterations of the angle deviation matrix and the signal matrix, the restored array received signal matrix is ​​obtained. 3) The ETAM algorithm and beamforming of the vector hydrophone array are then performed to estimate the target azimuth. This invention mainly utilizes the alternating iterations of the angle deviation matrix and the signal matrix to propose a DOA estimation method for moving vector hydrophone arrays with element angle deviations. Simulation results confirm that the proposed algorithm can improve the azimuth estimation performance and accuracy of the vector hydrophone array on the moving platform when there are element angle deviations. Attached Figure Description

[0097] Figure 1 It is the result of a single estimation of the bi-objective using the ETAM algorithm and the calibrated ETAM algorithm;

[0098] Figure 2 This is a graph showing the relationship between RMSE and SNR in the ETAM algorithm for dual-target azimuth estimation, which is used to calibrate the angular deviation of array elements.

[0099] Figure 3 The relationship between the RMSE of the dual-target azimuth estimation in the ETAM algorithm for calibrating the angular deviation of the array elements and the number of physical array elements;

[0100] Figure 4 This is the relationship between the RMSE of the dual-target azimuth estimation in the ETAM algorithm for calibrating array element angular deviation and the array element angular deviation.

[0101] Figure 5 This is the relationship between the RMSE of the dual-target azimuth estimation and the target spacing in the ETAM algorithm for calibrating the angular deviation of the array elements. Detailed Implementation

[0102] The vector hydrophones have 3 channels, a signal-to-noise ratio of 0dB, and target azimuths of -7° and 14°. A 4-element vector hydrophone linear array is used, and the array is synthesized twice, expanding by 2 elements each time, resulting in an 8-element vector array. A fixed bias of 10° and a random bias of 10° are added to each element. The received data matrix of the 4-element physical array has a dimension of 3000×12, where 3000 represents the total number of samples and 12 represents the total number of channels of the 4 vector hydrophone elements.

[0103] The principle of this invention is as follows: This invention combines a moving platform with a vector hydrophone synthetic aperture algorithm, and proposes a DOA estimation method for a moving vector hydrophone array with element angular deviations by iteratively using the angle deviation matrix and the signal matrix. This method first fixes the signal and optimizes the angle deviation matrix, then fixes the angle deviation matrix again and recovers the signal. Its main idea is to improve the azimuth estimation of a moving platform vector hydrophone array with angular deviations, thereby increasing the accuracy of the azimuth estimation. The results of this invention confirm that it can improve the azimuth performance and the accuracy of azimuth estimation.

[0104] This invention provides a method for estimating the DOA of a motion array in a vector hydrophone, the method comprising:

[0105] Step 1: Perform a coarse estimate of the angle deviation matrix G, then reconstruct the coarse estimate using the inherent directionality of the vector hydrophone to make the angle deviation matrix estimate more accurate. The coarse estimate of the angle deviation matrix is ​​written as:

[0106]

[0107] At this time The matrix structure does not satisfy expression (1), and the element magnitudes are not normalized. Let yes The normalized result, and Divide into H 2×2 sub-blocks, and represent the h-th block matrix as follows: Its expression is:

[0108]

[0109] Where α = 3h⁻¹, and It is related to the cosine value of the array element angle deviation. and It is related to the sine value of the angular deviation of the array elements. After performing the corresponding inverse cosine and arcsine transformations on each element, it is rewritten into the matrix, denoted as . Where i represents the i-th iteration, its expression is:

[0110]

[0111] To eliminate beam output deviation caused by array element angle misalignment, the relative deviation matrix of the nth vector hydrophone is first calculated, using the first array element as a reference. The relative deviation matrix is ​​denoted as... The expression is:

[0112]

[0113] make and These represent the angular deviation β obtained through the sine and cosine functions, respectively. n The estimated values ​​are expressed as follows:

[0114]

[0115]

[0116] However, simulation experiments revealed that if directly using... or This results in poor algorithm performance when estimating targets in the normal direction and end-fire direction. Therefore, the average weighting method is chosen to obtain the estimated angle deviation value of the nth vector hydrophone. Right now:

[0117]

[0118] At this point, the angle deviation matrix of the nth vector hydrophone is expressed as:

[0119]

[0120] Angular deviation matrix Reconstructed as:

[0121]

[0122] This completes the first step of the iterative algorithm.

[0123] The reconstructed angle deviation matrix has a dimension of 12×12, and the specific data it contains is as follows:

[0124] 4 0 0 0 0 0 0 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 0 0 1.7922 2.5062 0 0 0 0 0 0 0 0 0 0 -2.5062 1.7922 0 0 0 0 0 0 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 0 0 1.8113 2.3086 0 0 0 0 0 0 0 0 0 0 -2.3086 1.8113 0 0 0 0 0 0 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 0 0 3.8586 0.9613 0 0 0 0 0 0 0 0 0 0 -0.9613 3.8586

[0125] Step 2: Use the reconstructed angle deviation matrix from Step 1 By minimizing equation (12), the signal can be obtained. The estimated value is given by , where i is the i-th iteration. The derivation process omits the superscript of the iteration number and first sets...

[0126]

[0127] Where X = [x(1), ..., x(L)] T , for The function, Let P be an array manifold matrix, where P = [p1, p2, ..., p2]. M ] T The signal power matrix is ​​calculated using the following formula:

[0128] Based on the properties of the F-norm, the above equation It can be transformed into:

[0129]

[0130] Continue to solve Regarding signals Differential:

[0131]

[0132] Therefore, the function Regarding signals The expression for the first-order partial derivative is:

[0133]

[0134] make Signals can be obtained The estimation expression is:

[0135]

[0136] in:

[0137]

[0138] Here, R is the covariance matrix of the data received by the acoustic pressure hydrophone array.

[0139] Receive signal The estimated value is:

[0140]

[0141] Therefore, the signal power corresponding to each scan area becomes:

[0142]

[0143] in Finally, the weighted compensation factor ω is estimated. m Choosing ω by imitating the generalized information criterion m :

[0144]

[0145] Where L represents the number of snapshots within the measurement time, and N represents the number of array elements.

[0146] when When, under the generalized information criterion, expression (1) can be transformed into:

[0147]

[0148] make:

[0149]

[0150] Here, Γ is a parameter used in the process derivation.

[0151] If the compensation term in equation (11) is ignored, it is the same as the above equation. Therefore, according to expression (13), we can obtain:

[0152]

[0153] According to expression (20), the signal at this time is The estimated value is:

[0154]

[0155] Substituting equation (24) into equation (23), we get:

[0156]

[0157] Substituting equation (19) into the above equation, we get:

[0158]

[0159] Comparing equations (27) and (28), we find that R can be used directly. -1 Replace Q m , where R is the covariance matrix. Now, a compensation term is added, and let...

[0160]

[0161] Here These are parameters derived from the process.

[0162] in:

[0163]

[0164] Here These are parameters derived from the process.

[0165] The simplified version is obtained from equation (22). The expression is as follows:

[0166]

[0167] The dimension of the calibrated received data matrix remains 3000×12, and the dimension of its covariance matrix is ​​12×12.

[0168] The convergence condition for the alternating iteration of the angle deviation matrix and the signal matrix is: the difference between the angle deviation matrix and the identity matrix is ​​taken, and the loop can be exited if the 2-norm of the difference matrix is ​​less than 0.001.

[0169] Step 3: After alternating iterations of the angle deviation matrix and the signal matrix, the restored array received signal matrix is ​​obtained. Then, the vector hydrophone array ETAM algorithm and beamforming are performed to estimate the target azimuth.

[0170] The receiver matrix obtained after synthesizing the aperture using the ETAM algorithm has a dimension expansion of 1000×24, where 1000 represents the number of samples retained after synthesizing the aperture, and 24 represents the total number of channels in the synthesized 8 virtual vector array elements. The beamforming output is 1×1021, with 1201 columns because the scanning angle is from -30° to 30° and the interval is 0.05°.

[0171] This invention primarily focuses on the multi-target azimuth estimation and performance analysis of a vector hydrophone motion array DOA estimation method with element angular deviations. The experimental conditions are as follows: First, a 4-element vector hydrophone linear array with an element spacing of half a wavelength is used. The array is expanded twice, with two elements added each time, to synthesize an 8-element vector hydrophone array, which is required to distinguish two targets. The true azimuths of the targets are set to -7° and 14°, with an SNR of 10dB. A fixed 10° angular deviation and a random angular deviation within 10° are added to the array. The ETAM algorithm is selected for dual-target azimuth estimation. The signal frequency is set to 5kHz, the sampling frequency to 50kHz, the sound speed to 1500m / s, and the signal incident azimuth to 10°. Vector hydrophone arrays with an element spacing of half a wavelength are used, and 200 Monte Carlo experiments are performed for each group.

[0172] Figure 1 The results of a single-shot estimation of the dual target using the ETAM algorithm and the calibrated ETAM algorithm are shown. The estimation result of the ETAM algorithm is (-7.45°, 13.87°), while the estimation result of the calibrated ETAM algorithm is (-7.28°, 13.98°). It can be seen that the proposed algorithm can improve the accuracy of azimuth estimation and reduce the estimation error.

[0173] Figure 2 , Figure 3 , Figure 4 , Figure 5 The relationships between RMSE and SNR for dual-target azimuth estimation using the ETAM algorithm for calibrating array element angle deviations, the number of physical array elements, the array element angle deviations, and the target spacing are presented respectively.

[0174] in, Figure 2 The target azimuths were set to -7° and 14°, with a physical array element count of 4 (combined into an 8-element vector hydrophone array). A fixed 10° angle deviation and a random angle deviation within 10° were added to the array. The SNR ranged from 0dB to 20dB (in 5dB increments; too small an SNR would prevent matrix inversion in the algorithm). The RMSE of the dual-target azimuth estimation results using the ETAM algorithm based on a moving platform was obtained as a function of SNR. Figure 2 It can be seen that: 1) After increasing the element angle deviation, when the signal-to-noise ratio is below 15dB, the azimuth estimation performance of the uncalibrated ETAM algorithm drops sharply, but the algorithm proposed in this invention can still estimate the target azimuth, and after synthesizing an 8-element vector hydrophone array, the azimuth estimation RMSE is always below 0.5°. As the signal-to-noise ratio increases, the azimuth estimation accuracy improves slightly; 2) Overall, in the presence of element angle deviation, the dual-target azimuth estimation performance of the algorithm proposed in this invention is better than that of the uncalibrated ETAM algorithm.

[0175] Figure 3 The target azimuths were set to -7° and 14°, with an SNR of 10dB. A fixed 10° angle deviation and a random angle deviation within 10° were added to the array. The number of physical array elements ranged from 4 to 16 (with an interval of 2; an even number of elements facilitates expansion). The RMSE of the dual-target azimuth estimation results using the ETAM algorithm based on a moving platform was obtained as a function of the number of physical array elements. Figure 3 It can be seen that: 1) The more physical array elements there are, the better the performance of the ETAM algorithm and the calibrated algorithm; 2) Overall, in the case of array element angle deviation, the proposed algorithm has better azimuth estimation performance than the uncalibrated ETAM algorithm.

[0176] Figure 4 The target azimuths were set to -7° and 14°, with an SNR of 10dB and a physical array element count of 4 (combined into an 8-element vector hydrophone array). The element angle errors ranged from 2° to 20° (with intervals of 2). The RMSE of the dual-target azimuth estimation results using the ETAM algorithm based on a moving platform was obtained as a function of the element angle deviation. Figure 4 It can be seen that: 1) The larger the element angle deviation, the larger the azimuth estimation RMSE of the two algorithms; 2) When the element angle deviation is 2°, the azimuth estimation accuracy of the calibrated ETAM algorithm is only 0.1° higher than that of the original algorithm. When the angle deviation reaches 20°, the azimuth estimation accuracy of the calibrated ETAM algorithm is improved by 0.4°, indicating that the larger the deviation, the greater the impact of the calibration signal matrix on improving the azimuth estimation accuracy; 3) Overall, as long as there is an element angle deviation, the azimuth estimation accuracy of the proposed algorithm is better than that of the original ETAM algorithm.

[0177] Figure 5With an SNR of 10dB and a physical array element count of 4 (combined into an 8-element vector hydrophone array), a fixed 10° angular bias and a random angular bias within 10° were added to the array. The target spacing ranged from 14° to 30° (with a spacing of 2; for the expanded 8-element vector hydrophone array, its beamwidth is 12.5°, and the target spacing should be greater than 12.5° to distinguish between two targets). The trend of RMSE variation of the dual-target azimuth estimation results using the ETAM algorithm based on a moving platform with the target spacing was obtained. Figure 5 It can be seen that: 1) Under the premise that the algorithm can distinguish two targets, the larger the target interval, the smaller the azimuth estimation RMSE. The estimation accuracy of the uncalibrated ETAM algorithm tends to be stable when the target interval reaches 26°, and the calibrated ETAM algorithm is basically stable when the target interval is 20°; 2) Overall, as long as there is an element angle deviation, the azimuth estimation accuracy of the algorithm proposed in this invention is better than that of the original ETAM algorithm.

Claims

1. A method for estimating the DOA of a motion array in a vector hydrophone, characterized in that: The vector hydrophone motion array DOA estimation method includes the following steps: 1) Obtain the element angle deviation matrix and perform a coarse estimation of the element angle deviation matrix; 2) Reconstruct the coarse estimation result of the array element angle deviation matrix obtained in step 1) to obtain the reconstructed angle deviation matrix; 3) Calculate the estimated value of the signal matrix from the angle deviation matrix obtained in step 2); iterate the estimated value of the signal matrix and the reconstructed angle deviation matrix obtained in step 2) alternately to obtain the restored array received signal matrix; 4) Based on the restored array received signal matrix obtained in step 3), perform the vector hydrophone array ETAM algorithm and beamforming to estimate the target azimuth.

2. The vector hydrophone motion array DOA estimation method according to claim 1, characterized in that: In step 1), the array element angle deviation matrix is ​​G, and the form of the array element angle deviation matrix G is as follows: The result of the coarse estimation of the pair element angle deviation matrix is The The expression is: in: g is the angular deviation matrix of the vector hydrophone; N is the total number of array elements; X is the data matrix received by the vector hydrophone array; It is an array manifold matrix; It is the target signal vector.

3. The vector hydrophone motion array DOA estimation method according to claim 2, characterized in that: The specific implementation method of step 2) is as follows: 2.1) Results of coarse estimation of the pair element angle deviation matrix Normalization is performed to obtain the normalized result. ; 2.2) Normalize the result obtained in step 2.1). Divided into indivual The sub-block, of which the first Each sub-block is represented as Where h ≤ H; the The expression is: in: ； and It is related to the cosine value of the array element angle deviation; and It is related to the sine value of the array element angle deviation; It is the α-th row and α-th column of the normalized result; It represents the α+1th row and α+1th column of the normalized result; yes The normalization result; 2.3) After performing the corresponding inverse cosine and arcsine transformations on each element of the pairwise angle deviation matrix, rewrite the matrix, denoted as . Where i represents the i-th iteration, the The expression is: 2.4) Taking the first element in the element angle deviation matrix as a reference, first solve for the... The relative deviation matrix of a vector hydrophone, denoted as [missing information] The The expression is: ； Where: Re is the real part of the complex number; 2.5) The relative deviation matrix of the vector hydrophone obtained in step 2.4). Based on this, construct the angle deviation matrix of the vector hydrophone. The angular deviation matrix of the vector hydrophone The expression is: in: It is the first Angle deviation estimates for each vector hydrophone; The The expression is: in: as well as The angular deviations are obtained using the sine and cosine functions, respectively. The estimated value; The The expression is: The The expression is: ； 2.6) The angular deviation matrix of the vector hydrophone obtained in step 2.5). The angle deviation matrix of the basic pair is reconstructed to obtain the reconstructed angle deviation matrix. The reconstructed angle deviation matrix The expression is: in: n≤N.

4. The vector hydrophone motion array DOA estimation method according to claim 3, characterized in that: The specific implementation method of step 3) is as follows: 3.1) Obtaining the signal The It is a signal matrix. The m-th vector in the middle; 3.2) Based on the reconstructed angle deviation matrix obtained in step 2), For the signal obtained in step 3.1) Make predictions and obtain signals The estimated value ; 3.3) The signal estimate and the angle deviation matrix are iterated alternately.

5. The vector hydrophone motion array DOA estimation method according to claim 4, characterized in that: The signal in step 3.1) The expression is: in: X represents the data matrix received by the vector hydrophone array; T stands for transpose; ； R is the covariance matrix of the data received by the acoustic bar hydrophone array; It is an array manifold matrix; H is the conjugate transpose, which is the conjugate transpose of A; pm represents the m-th row in the signal power matrix P; the signal power matrix is ; Where m is 1, 2, 3, ..., M; L represents the number of snapshots within the measurement time; ； i represents the i-th iteration.

6. The vector hydrophone motion array DOA estimation method according to claim 5, characterized in that: The signal in step 3.2) The estimated value The expression is: in: It is a weighted compensation factor; The expression is: in: L represents the number of snapshots within the measurement time; N represents the number of elements in the vector hydrophone array; It is the estimated value of pm in the i-th iteration; These are process derivation parameters, the stated The expression is: in: X represents the data matrix received by the vector hydrophone array; Tr represents the trace of the matrix; These are process derivation parameters, the stated The expression is: 。 7. The vector hydrophone motion array DOA estimation method according to claim 6, characterized in that: The signal power corresponding to the scanning area in step 3.3) is The The expression is: in: It is an iteration of the signal, and its expression is: .

8. The vector hydrophone motion array DOA estimation method according to claim 7, characterized in that: The convergence condition for the iteration in step 3.3) is: the difference between the angle deviation matrix and the identity matrix is ​​taken, and if the second norm of the difference matrix is ​​less than 0.001, the iteration loop is exited.

Images