Subarray beam domain high-resolution spatial spectrum estimation method under undersnapshot condition

Abstract

The invention discloses a subarray beam domain high-resolution spatial spectrum estimation method under an under-snapshot condition. The method comprises the following steps: step 1, uniformly dividing N-element towed linear array sonar array element frequency domain data into K subarrays; 2, performing beam forming processing on each piece of sub-array data to obtain a beam domain result of each sub-array; 3, calculating a beam domain steering vector; 4, calculating a beam domain covariance matrix; 5, realizing signal and noise power parameter estimation based on a sparse approximate minimum variance criterion; and 6, carrying out frequency band weighted summation on the estimated signal power to obtain a broadband DOA estimation result. According to the method, sub-array division dimensionality reduction, beam domain coherent processing and sparse parameter iteration estimation are combined for processing, the method has the advantages of low snapshot adaptability and high anti-interference capability, the method can be used for near-field weak signal detection of the towed linear array under multi-target interference, and effectiveness is verified through marine test data.

Description

Technical Field

[0001] This invention belongs to the field of sonar signal processing, specifically to the research field of passive sonar weak signal detection technology, and in particular to a subarray beam domain high-resolution spatial spectrum estimation method under low snapshot conditions. Background Technology

[0002] Towed linear array sonar, due to its large aperture and low operating frequency, has gradually become the main sonar for underwater anti-submarine warfare on Chinese ships and vessels. With the breakthrough development of advanced submarine acoustic stealth technology, towed linear arrays need to continuously increase their array aperture to improve detection gain in order to maintain their detection capabilities. However, with the increase in aperture and the change in detection targets from the far field to the near field, traditional high-resolution detection algorithms (such as MUSIC and ESPRIT) face new challenges, exhibiting phenomena such as insufficient effective snapshots, inability to converge the covariance matrix, a sharp decline in interference suppression performance, and the signal being easily mistaken for interference. Therefore, developing high-resolution spatial spectrum estimation methods for subarray beam domains under under-snapshot conditions has become increasingly important. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a high-resolution spatial spectrum estimation method for subarray beam domain under low-shot conditions, which better adapts to low-shot conditions, reduces dependence on data volume, and also has the advantages of strong anti-interference ability and good near-field weak signal detection effect of towed linear arrays under multi-target interference.

[0004] The technical solution of this invention is as follows:

[0005] A high-resolution spatial spectrum estimation method for subarray beam domain under under-shot conditions includes:

[0006] Step 1: Array partitioning, The element frequency domain data of the towed linear array sonar array are uniformly divided into Individual formation;

[0007] Step 2: Subarray beamforming. Beamforming is performed on the data of each subarray to obtain the beam domain results for each subarray.

[0008] Step 3: Calculate the beam domain steering vector;

[0009] Step 4: Calculate the beam domain covariance matrix;

[0010] Step 5: Estimate signal and noise power parameters based on the sparse approximate minimum variance criterion;

[0011] Step 6: Perform band-weighted summation on the estimated signal power to obtain the broadband DOA estimation result.

[0012] The method of this invention starts from the direction of high-resolution spatial spectrum estimation technology based on sparse parameter estimation. It combines subarray partitioning and dimensionality reduction, beam domain coherent processing and sparse parameter iterative estimation to effectively solve the problem of near-field detection of towed array sonar under conditions of few snapshots. Its core advantages are: (1) Adaptability to few snapshots: through sparse parameter iterative estimation and subarray beam domain dimensionality reduction processing, the dependence on data volume is reduced and the convergence speed of parameter estimation is accelerated; (2) Strong anti-interference capability: the beam domain can naturally suppress non-target azimuth interference, and the sparse approximate minimum variance iteration further optimizes the sparsity. It can be applied to the near-field weak signal detection of towed linear arrays under multi-target interference, and its effectiveness has been verified by sea test data. Attached Figure Description

[0013] Figure 1 This is a flowchart of the workflow of the present invention;

[0014] Figure 2 The simulation results are for a drag line array (128 elements, 3m spacing);

[0015] Figure 3 Simulation results for another towed line array (128 elements, 3m spacing);

[0016] Figure 4a , Figure 4b , Figure 4c , Figure 4d The data processing results for a certain towed linear array (120 elements, 1.5m spacing) during sea trials are as follows: Figure 4a The trajectory diagram for the CBF method. Figure 4b The trajectory diagram for the MVDR method. Figure 4c The trajectory diagram of the SubBeamSAMV method. Figure 4d This is a comparison chart of the snapshot results from the three methods. Detailed Implementation

[0017] The present invention will be further described below with reference to specific embodiments and accompanying drawings:

[0018] This invention provides a high-resolution spatial spectrum estimation method for subarray beam domain under under-shot conditions, such as... Figure 1 As shown, the specific steps are as follows:

[0019] (1) Array partitioning, Element-level towed linear array sonar array element frequency domain data (expression: ) evenly divided into Each subarray contains [number] subarrays. Each array element, The subarray data matrix is ​​as follows:

[0020]

[0021] in, To handle frequency, and To handle the upper and lower limits of frequency, For the subarray element index, Subarray number;

[0022] (2) Subarray beamforming: Beamforming is performed on the data of each subarray to obtain the beam domain results of each subarray;

[0023] First, calculate the manifold matrix of the subarray array:

[0024]

[0025] in, For the spacing between array elements, For the speed of sound, The imaginary unit, M is the subarray number, and M is the number of elements in the subarray. The subarray spatial scanning orientation, total number of scans indivual;

[0026] Then, subarray beam domain data is generated based on the subarray manifold matrix:

[0027]

[0028] in, For the first All array metadata of each subarray;

[0029] Merge all subarray beam output data:

[0030] ;

[0031] (3) Calculate the beam domain steering vector.

[0032] First, calculate the array manifold matrix of the entire array:

[0033]

[0034] in, To handle frequency, For the total number of elements in the entire array, For the speed of sound, The imaginary unit, The total number of scans is the spatial scanning orientation of the entire array. indivual( ),

[0035] The beam domain steering vector is then calculated as follows:

[0036]

[0037] in, For the sub-array Array manifold matrix, For subarray The full array manifold corresponding to the position of each array element. To handle frequency, For subarray numbers, The total number of subarrays, This indicates the azimuth of the entire array spatial scan. The subarray spatial scanning orientation, total number of scans indivual;

[0038] (4) Calculate the beam domain covariance matrix:

[0039]

[0040] in, To handle frequency, The subarray spatial scanning orientation;

[0041] (5) Signal and noise power parameter estimation is achieved based on the sparse approximate minimum variance criterion.

[0042] Repeat for each frequency cycle:

[0043] ① Set the initial signal power With noise power :

[0044] ,

[0045] in, To find the trace of a matrix, The total number of subarrays, This represents the total number of scans.

[0046] ② Iteratively update signal power With noise power ( (Number of iterations)

[0047] Construct the covariance matrix.

[0048]

[0049] in, As a unit array, ,

[0050] Update signal and noise power.

[0051]

[0052]

[0053] ③ The update iteration ends, and the final signal estimate is obtained. .

[0054] (6) Estimated signal power The broadband DOA estimation result is obtained by performing band-weighted summation. .

[0055] Figure 2 The simulation results are for a towed linear array with parameters of 128 elements, 3m spacing, 250Hz processing frequency, 8 subarrays, 16 data snapshots, and simulated target azimuths of [40, 45, 90, 120, 130]°, and signal-to-noise ratios of [-15, 5, 3, 2, -18] dB. Comparing the results of conventional processing (CBF), traditional adaptive processing (MVDR-DL), and the proposed method, it can be observed that other methods suffer from severe sidelobe leakage due to strong interference. For a 40° weak signal, conventional detection algorithms mask the weak signal due to severe leakage at 45°. While the traditional adaptive detection method suppresses sidelobe leakage due to strong interference, the covariance matrix cannot converge properly due to the limited number of snapshots. The diagonal loading method used in this method fails to detect the weak signal. The proposed method can quickly detect weak signals under strong interference conditions with few snapshots, and its weak signal detection performance is the best among the three algorithms.

[0056] Figure 3 The simulation results for another draggable linear array are compared with... Figure 2 Under the same array parameters, processing parameters, and other conditions, and with an improved interference signal power ratio, the simulated target azimuth is [40, 45, 90, 120]°, and the signal-to-noise ratio is [-18, 7, -3, -2] dB. Comparing the detection results of the three algorithms, it can be found that under strong interference background, the detection performance of the method of this invention is still the best for detecting weak signals at 40°.

[0057] Figure 4a , Figure 4b , Figure 4c , Figure 4d This is the data processing result of a sea trial of a towed linear array. The parameters are 120 elements, 1.5m spacing, processing frequency band 100-150Hz, 24 batches of valid snapshot data, 8 subarrays processed, and 16 subarray elements. Figure 4a The trajectory diagram for the CBF method. Figure 4b The trajectory diagram for the MVDR method. Figure 4c The trajectory diagram of the SubBeamSAMV method. Figure 4d This is a comparison chart of the snapshot results from the three methods.

[0058] Comparing the results of conventional processing (CBF), traditional adaptive processing (MVDR-DL), and the processing of this invention (where the loading amount of MVDR-DL is 0.1 times the mean of the diagonal elements), it can be found that MVDR-DL is severely affected by missing data, the target energy is intermittent, and it cannot detect weak targets near 70°. CBF is severely affected by interference leakage, and its weak target detection capability is weak. The SubBeamSAMV method of this invention can converge quickly, the trajectory of weak targets is clearly visible, and its detection capability is significantly better than the other two algorithms.

[0059] In summary, the proposed method for high-resolution spatial spectrum estimation in the subarray beam domain under low-shot conditions for towed linear arrays firstly divides the large-scale linear array into multiple non-overlapping subarrays, reducing data dimensionality and improving system robustness. Then, each subarray independently performs beamforming, transforming element-domain data into beam-domain data and suppressing the influence of noise and non-target azimuth interference. Finally, based on the sparsity characteristics of the target signal in the spatial domain, the sparse approximate minimum variance criterion is used to achieve iterative optimization estimation of beam-domain signal and noise power. Verification using sea trial data shows that this method effectively solves the robust estimation problem of the covariance matrix under low-shot conditions and can significantly improve the detection performance of towed array sonar under these conditions.

[0060] It should be noted that the above embodiments are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Equivalent modifications made based on the above embodiments are all within the scope of protection of the present invention.

Claims

1. A high-resolution spatial spectrum estimation method for subarray beam domain under under-shot conditions, characterized in that, include: Step 1: Array partitioning, The element frequency domain data of the towed linear array sonar array are uniformly divided into Individual formation; Step 2: Subarray beamforming. Beamforming is performed on the data of each subarray to obtain the beam domain results for each subarray. Step 3: Calculate the beam domain steering vector; Step 4: Calculate the beam domain covariance matrix; Step 5: Estimate signal and noise power parameters based on the sparse approximate minimum variance criterion; Step 6: Perform band-weighted summation on the estimated signal power to obtain the broadband DOA estimation result.

2. The subarray beam domain high-resolution spatial spectrum estimation method under under-shot conditions according to claim 1, characterized in that: In step one, The expression for the element frequency domain data of a towed linear array sonar array is: ； Divided into After each subarray, each subarray contains Each array element, The expression for the submatrix data matrix is: ； in, To handle frequency, and To handle the upper and lower limits of frequency, For the subarray element index, This is the subarray number.

3. The subarray beam domain high-resolution spatial spectrum estimation method under under-shot conditions according to claim 2, characterized in that: In step two, beamforming processing is performed on the data of each subarray to obtain the beam domain result of each subarray, then: First, calculate the manifold matrix of the subarray array: ； in, For the spacing between array elements, For the speed of sound, The imaginary unit, M is the subarray number, and M is the number of elements in the subarray. The subarray spatial scanning orientation, total number of scans indivual; Then, subarray beam domain data is generated based on the subarray manifold matrix: ； in, For the first All array metadata for each subarray; Merge all subarray beam output data: 。 4. The subarray beam domain high-resolution spatial spectrum estimation method under under-shot conditions according to claim 3, characterized in that: In step three, the process of calculating the beam domain steering vector is as follows: First, calculate the array manifold matrix of the entire array: ； in, To handle frequency, For the total number of elements in the entire array, For the speed of sound, The imaginary unit, The total number of scans is the spatial scanning orientation of the entire array. indivual( ), The beam domain steering vector is then calculated as follows: ； in, For the sub-array array manifold matrix, For subarray The full array manifold corresponding to the position of each array element. To handle frequency, For subarray numbers, The total number of subarrays, This indicates the azimuth of the entire array spatial scan. The subarray spatial scanning orientation, total number of scans indivual.

5. The subarray beam domain high-resolution spatial spectrum estimation method under under-shot conditions according to claim 4, characterized in that: In step four, the expression for calculating the beam domain covariance matrix is ​​as follows: ； in, To handle frequency, This indicates the spatial scanning orientation of the subarray.

6. The subarray beam domain high-resolution spatial spectrum estimation method under under-shot conditions according to claim 5, characterized in that: In step five, the process of estimating signal and noise power parameters based on the sparse approximate minimum variance criterion is as follows: Repeat for each frequency cycle: ① Set the initial signal power With noise power : ， ； in, To find the trace of a matrix, The total number of subarrays, This represents the total number of scans. ② Iteratively update signal power With noise power ( (Number of iterations) Construct the covariance matrix. ； in, As a unit array, , Update signal and noise power. ； ； ③ The update iteration ends, and the final signal estimate is obtained. .

7. The subarray beam domain high-resolution spatial spectrum estimation method under under-shot conditions according to claim 6, characterized in that: In step six, the estimated signal power is subjected to band-weighted summation to obtain the expression for the broadband DOA estimation result: .

Images