Research On Underwater DOA Estimation Algorithm Based On Nested Array

Direction of Arrival(DOA)estimation is a basic problem in array signal processing.Underwater DOA estimation is widely used in underwater navigation,underwater acoustic positioning,ocean mining and other fields.In the actual underwater environment,sound waves are often used for propagation,and the velocity of sound waves changes with the change of environmental factors.The traditional DOA estimation algorithm calculates the sound velocity as a constant,which will bring large estimation deviation.However,the existing underwater DOA algorithm uses uniform linear array,so the estimation accuracy is not high enough.This paper mainly studies the velocity-independent underwater DOA estimation algorithm based on the nested array.The main work includes:(1)The influence of sound velocity on DOA estimation is studied,and the concept of root mean square error(₍(8(8))of sound velocity correlation is proposed,which is proved to be the upper performance limit of traditional DOA estimation algorithm in theory and simulation.It provides theoretical guidance for the research of underwater velocity-independent DOA estimation algorithm.(2)A new array structure,cross-nested array,is proposed.The cross-nested array is composed of several nested arrays and has the advantages of both nested arrays and cross uniform linear arrays.Compared with the cross uniform linear array,the cross-nested array has a large enough antenna aperture to improve the estimation accuracy.Compared with the L-shaped nested arrays,the crossover structure of the cross-nested array can eliminate the error caused by inaccurate sound velocity.(3)A one-dimensional underwater DOA estimation algorithm based on two-dimensional cross-nested array(1D-SSVI-ESPRIT),which is independent of sound velocity,is proposed.The proposed algorithm performs one-dimensional angle estimation on the two nested arrays,and establishes a relationship between the two sets of estimated values based on the special crossing angle.The estimated values are paired by sorting,and the paired values are divided to eliminate the velocity variable.Thus,the velocity-independent one-dimensional DOA estimation results are obtained,and the real-time sound velocity estimation can also be obtained.Compared with the traditional algorithm based on nested array,the estimation result of the proposed algorithm is velocity-independent;compared with the existing one-dimensional velocity-independent DOA algorithm,the proposed algorithm has higher estimation accuracy and better robustness.(4)A two-dimensional underwater DOA estimation algorithm based on three-dimensional cross-nested array(2D-SSVI-ESPRIT)is proposed.The proposed algorithm estimates the angle of three nested arrays by one-dimensional angle estimation,traverses all pairing combinations,combines the sound velocity range and obtains the sound velocity estimation according to the principle of minimum variance of sliding window.The velocity value is used to correct the previous three groups of angles and pairing,so as to obtain the velocity-independent two-dimensional DOA estimation results.Compared with the two-dimensional DOA estimation algorithm based on L-type nested array,the estimation result of the proposed algorithm is velocity-independent;compared with the existing two-dimensional velocity-independent DOA algorithm,the proposed algorithm has higher estimation accuracy and better robustness.