CN-121978694-A - Underwater target detection and positioning method and system based on passive sonar array

CN121978694ACN 121978694 ACN121978694 ACN 121978694ACN-121978694-A

Abstract

The invention discloses an underwater target detection and positioning method and system based on a passive sonar array, which relate to the technical field of underwater sound signal processing and are used for receiving acoustic signals radiated by an underwater target and dividing the acoustic signals into dense subarray data and sparse subarray data, operating a sparse Bayesian learning algorithm on the sparse subarray data to output a high-precision fuzzy value, carrying out azimuth estimation on the dense subarray data in a full-angle range through a dense part to obtain a low-precision non-fuzzy value, and fusing the high-precision fuzzy value and the low-precision non-fuzzy value through an estimation matching model to obtain an azimuth estimation result of the underwater target. The method carries out high-precision azimuth estimation based on the combination of the ternary mutual mass array and the compressed grid sparse Bayesian learning, carries out high-precision (but fuzzy) estimation by using the sparse part of the TCA array, carries out low-precision (but non-fuzzy) estimation by using the dense part, and realizes high-precision target azimuth estimation under low operation quantity.

Inventors

DU YANGFAN
ZHAO XUANZHI
DUAN HUIFANG
CHEN ZHENG
LIU ZENGLI

Assignees

昆明理工大学

Dates

Publication Date: 20260505
Application Date: 20260123

Claims (8)

1. The underwater target detection and positioning method based on the passive sonar array is characterized by comprising the steps of constructing a ternary mutual mass array, wherein the ternary mutual mass array comprises a dense part and a sparse part; receiving an acoustic signal radiated by an underwater target, and dividing the acoustic signal into dense subarray data and sparse subarray data; analyzing an angle fuzzy mechanism of the sparse part according to a sampling theorem to construct a compressed grid; based on the compression grid, a sparse Bayesian learning algorithm is operated on the sparse subarray data, and a high-precision fuzzy value of azimuth estimation is output; Carrying out azimuth estimation on the dense subarray data in the full-angle range through the dense part to obtain a low-precision unambiguous value of azimuth estimation; And constructing an estimated matching model, and fusing a high-precision fuzzy value and a low-precision non-fuzzy value of the azimuth estimation through the estimated matching model to obtain an azimuth estimation result of the underwater target.
2. The method for detecting and locating an underwater target based on a passive sonar array according to claim 1, wherein said dense portion adopts a prototype mutual mass array for providing a low-precision unambiguous estimation of the azimuth; The sparse part adopts a sparse uniform subarray and is used for providing high-precision fuzzy estimation of the azimuth.
3. The method for detecting and locating an underwater target based on a passive sonar array according to claim 1, wherein the analyzing the angle ambiguity mechanism of the sparse part by using the sampling theorem to construct the compressed grid comprises: Calculating the subinterval width of the non-repeated normalized spatial frequency corresponding to the sparse subarray data; Dividing grid points within the subintervals.
4. The method for detecting and locating an underwater target based on a passive sonar array according to claim 1, wherein based on the compression grid, a sparse bayesian learning algorithm is operated on the sparse subarray data to output a high-precision fuzzy value of azimuth estimation, comprising: The sparse subarray data is taken as input, a complete dictionary matrix is constructed based on a compression grid, posterior mean and posterior covariance are calculated through a sparse Bayesian learning algorithm, super parameters are updated and are converged in an iterative mode, and a high-precision fuzzy value of an aliasing version of the true target azimuth is output.
5. The underwater target detection and positioning method based on the passive sonar array according to claim 2, wherein the ternary mutual mass array comprises a first subarray, a second subarray and a third subarray; the first subarray comprises M array elements with a spacing of N D, d is half wavelength; The second subarray comprises N array elements with the interval of M d; The first subarray and the second subarray form a prototype mutual mass array; the third subarray comprises Q array elements with a spacing of L D, where l=m+n, as a sparse portion.
6. The method for detecting and locating an underwater target based on a passive sonar array according to claim 1, wherein the method for estimating the azimuth of the dense subarray data in the full angle range by the dense part to obtain a low-precision unambiguous value of the azimuth estimation comprises the following steps: Constructing a grid in a full angle range; carrying out azimuth estimation on the dense subarray data based on the constructed grid; And outputting a low-precision non-fuzzy value set of azimuth estimation after algorithm iteration convergence, wherein each estimated value corresponds to a real interval range of the azimuth of the underwater target.
7. The method for detecting and locating an underwater target based on a passive sonar array according to claim 1, wherein the step of obtaining the result of estimating the underwater target by fusing the high-precision fuzzy value and the low-precision non-fuzzy value of the azimuth estimation by the estimation matching model comprises the steps of: Aiming at each low-precision non-fuzzy value, according to the geometric characteristics of the sparse subarray and the array element spacing parameters, calculating all aliasing azimuth sets corresponding to each low-precision non-fuzzy value by combining a space sampling theorem; Aiming at each high-precision fuzzy value, calculating Euclidean distance between each high-precision fuzzy value and each element in the aliasing azimuth set; Selecting a high-precision fuzzy value with the minimum Euclidean distance as a matching result; reversely mapping the matching result back to the full-angle space according to the angle aliasing mechanism of the sparse subarray to obtain a real high-precision azimuth; And integrating all the true high-precision orientations obtained by matching to form a final underwater target orientation estimation result set.
8. The underwater target detection and positioning system based on the passive sonar array is characterized by comprising a ternary mutual mass array construction module, a detection module and a positioning module, wherein the ternary mutual mass array construction module is used for constructing a ternary mutual mass array, and the ternary mutual mass array comprises a dense part and a sparse part; the signal receiving module is used for receiving acoustic signals radiated by an underwater target and dividing the acoustic signals into dense subarray data and sparse subarray data through a ternary mutual mass array; the analysis module is used for analyzing the angle fuzzy mechanism of the sparse part according to the sampling theorem and constructing a compressed grid; The high-precision fuzzy value output module is used for operating a sparse Bayesian learning algorithm on the sparse subarray data based on the compression grid and outputting a high-precision fuzzy value of azimuth estimation; the low-precision non-fuzzy value output module is used for carrying out azimuth estimation on the dense subarray data in the full-angle range through the dense part to obtain a low-precision non-fuzzy value of azimuth estimation; and the structure fusion output module is used for constructing an estimated matching model, and the high-precision fuzzy value and the low-precision non-fuzzy value of the azimuth estimation are fused through the estimated matching model to obtain the azimuth estimation result of the underwater target.

Description

Underwater target detection and positioning method and system based on passive sonar array Technical Field The invention relates to the technical field of underwater acoustic signal processing, in particular to an underwater target detection and positioning method and system based on a passive sonar array. Background Passive sonar is a key technology for detecting, identifying and positioning underwater targets, and under far-field assumption, the traditional method for improving the estimation accuracy of the linear array azimuth is to increase the array aperture. However, conventional half-wavelength pitch Uniform Linear Arrays (ULA) are limited by the number of physical array elements, have limited resolution and degrees of freedom (DOFs), and are costly at large apertures. To break through this limitation, sparse arrays (SPARSE LINEAR ARRAY, SLA) such as nested arrays (NESTED ARRAY) and mutual mass arrays (Coprime Array) have been developed. They achieve larger virtual pore sizes and degrees of freedom through non-uniform alignment. However, the existing sparse array still has the problem of strong mutual coupling effect. In terms of azimuth estimation algorithm, traditional beam forming (CBF) resolution is low, MVDR and other adaptive methods rely on accurate estimation of covariance matrix, and the performance of the method is severely degraded in underwater complex dynamic environments (such as limited snapshot numbers). Sparse Bayesian Learning (SBL) is an emerging algorithm that performs well in low signal-to-noise ratio and coherent source environments. However, SBL suffers from "grid mismatch" in that increasing grid density reduces errors but increases computation dramatically and results in unstable values, whereas existing Off-grid methods are computationally complex. Particularly, in a few snapshot scene, how to realize high-precision and unambiguous azimuth estimation without changing the physical array structure is a technical problem to be solved. Therefore, aiming at the defects of the prior art, how to provide an underwater target detection and positioning method and system based on a passive sonar array, which can increase the degree of freedom while keeping low mutual coupling, and realize high-precision and non-fuzzy azimuth estimation without changing the physical array structure under a few snapshot scene is a problem to be solved by the technicians in the field. Disclosure of Invention In view of the above, the invention provides an underwater target detection and positioning method and system based on a passive sonar array, which overcome the problems of insufficient resolution, high computational complexity and grid flap ambiguity of a sparse array in a few-snapshot dynamic underwater sound environment in the prior art, perform high-precision azimuth estimation based on a ternary mutual mass array (Ternary Coprime Array, TCA) combined with compressed grid sparse Bayesian learning (CG-SBL), perform high-precision (but ambiguous) estimation by using a sparse part of the TCA array, perform low-precision (but unambiguous) estimation by using a dense part, and realize high-precision target azimuth estimation under low operation amount by compressing a grid strategy and estimating a matching frame. In order to achieve the aim, the invention adopts the following technical scheme that the underwater target detection and positioning method based on the passive sonar array comprises the steps of constructing a ternary mutual mass array, wherein the ternary mutual mass array comprises a dense part and a sparse part; Receiving an acoustic signal radiated by an underwater target, and dividing the acoustic signal into dense subarray data and sparse subarray data through a ternary mutual mass array; analyzing an angle fuzzy mechanism of the sparse part according to a sampling theorem to construct a compressed grid; based on the compression grid, a sparse Bayesian learning algorithm is operated on the sparse subarray data, and a high-precision fuzzy value of azimuth estimation is output; Carrying out azimuth estimation on the dense subarray data in the full-angle range through the dense part to obtain a low-precision unambiguous value of azimuth estimation; And constructing an estimated matching model, and fusing a high-precision fuzzy value and a low-precision non-fuzzy value of the azimuth estimation through the estimated matching model to obtain an azimuth estimation result of the underwater target. Preferably, the dense part adopts a prototype mutual mass array for providing low-precision unambiguous estimation of the azimuth; The sparse part adopts a sparse uniform subarray and is used for providing high-precision fuzzy estimation of the azimuth. Preferably, the analysis of the angle ambiguity mechanism of the sparse part by using the sampling theorem, the construction of the compressed grid, includes: Calculating the subinterval width of the non-repeated normalized spatial frequency corres