DOA Estimation Method For Vector Hydrophone Array Based On Sparse Representation

Direction of Arrival(DOA)estimation is an important research direction in vector hydrophone array signal processing.In recent years,with the rapid development of Sparse Representation(SR)theory,DOA estimation methods of vector hydrophone array based on sparse representation have become a research focus in modern signal processing field.Although sparse representation theory has shown better performance than traditional subspace algorithms in DOA estimation of vector hydrophone arrays,there are still some problems to be solved:(1)The DOA estimation algorithm for vector hydrophone array based on greedy type sparse signal recovery algorithm has a low efficiency in the selection process of atoms and does not correct the selected atoms.When facing complex signal data,the algorithm has weak processing capacity and noise resistance.(2)In the on-grid sparse representation of the vector hydrophone array DOA estimation algorithm,the spatial domain is discretized into a finite number of grids.However,the actual incidence angle of the signal in practical applications often does not fall precisely on the pre-defined grids,resulting in a grid mismatch phenomenon.The gridless sparse representation method is based on the atomic norm theory and does not require grid partitioning.This method avoids the problem of grid mismatch and converts the reconstruction problem of the sparse signal into a semi-definite programming(SDP)problem by constructing a continuous atomic set.It can obtain more accurate DOA estimation results without grid search.However,this type of method has not yet been applied to vector hydrophone array DOA estimation.This thesis is based on Sparse Representation theory and focuses on the DOA estimation method of vector hydrophone arrays.The main work is as follows:(1)Aiming at the low estimation accuracy and long processing time of the greedy type sparse reconstruction algorithm,such as Orthogonal Matching Pursuit(OMP),Compressive Sampling Matching Pursuit(Co Sa MP),and Subspace Pursuit(SP),in the on-grid sparse representation of the vector hydrophone array DOA estimation method,a Detouring Matching Pursuit(DMP)algorithm based method is proposed.It effectively improves the accuracy of atom selection,the accuracy of sparse signal recovery,and maintains high computational efficiency.Simulation results show that under the background of small snapshots and low signal-to-noise ratio,the proposed method in this thesis can achieve higher estimation accuracy compared to traditional subspace-based DOA estimation methods and sparse representation DOA estimation methods based on orthogonal matching pursuit.It can also achieve accurate DOA estimation in the case of coherent sources.(2)Aiming at the grid mismatch problem in the on-grid sparse representation of the vector hydrophone array DOA estimation problem,a gridless sparse representation vector hydrophone array DOA estimation method based on atomic norm minimization(ANM)is proposed,which has achieved obvious effects in solving the grid mismatch problem.For multi-snapshots situations,to reduce computational complexity,singular value decomposition(SVD)is used to reduce the dimensionality of the array reception data,and the alternating direction method of multipliers(ADMM)is used to solve the SDP problem to improve computational efficiency.This achieves efficient reconstruction of the Toeplitz matrix of the array reception signal.Finally,the root-MUSIC algorithm,extended to vector hydrophone arrays,is used to do the vandermonde decomposition of Toeplitz covariance matrix,which successfully estimates the DOA of the target signal.Simulation results demonstrate that,in the context of grid mismatch,the proposed method in this thesis achieves higher estimation accuracy than the on-grid sparse representation DOA estimation algorithm.