Research On DOA Estimation Method Based On Symmetric Characteristics Of Steering Vector

The azimuth estimation of underwater targets is an important research content in the field of array signal processing and underwater target detection and recognition.It has been the focus of many domestic and foreign scholars,and it is of great significance to achieve fast and efficient DOA estimation of underwater targets.In order to achieve higher estimation accuracy,in the spectrum search,measures to reduce the step size of the search are often taken,which will increase the amount of computation sharply.This paper starts with reducing the amount of spectral search and proposes a method for MVDR algorithm.A method of fast spatial spectrum search.By studying the uniform linear array,we can find that there is a symmetric relationship between the array steering vectors.Based on this symmetric relationship,an improved minimum variance undistorted response algorithm is proposed based on the MVDR algorithm.The spectral function of the algorithm contains the source orientation information of the entire spatial spectrum in half of the spatial spectrum,and the source orientation information in the half spatial spectrum is determined to obtain the true angle information of the source,thereby reducing the spectral search amount by 50.%.Compared with the calculation of the MVDR algorithm under different array elements and different spatial angle search steps,the calculations of both the number of array elements and the reduction of spatial angle search step are gradually increased.Relatively speaking,the computational step size of the reduced space angle is sharply increased,and the improved MVDR algorithm has an advantage in estimating efficiency.Aiming at the phenomenon that the spectral search of the improved MVDR algorithm is easy to cause the peak of the true source and the mirror source,the method of quadratic local search at the spatial peak position and the symmetrical position of the improved MVDR algorithm can achieve accurate orientation..Although some calculations are added,it is possible to save more computational time in achieving comparable DOA estimation accuracy than the MVDR algorithm.The uniform vector linear array steering vector is different from the sound pressure matrix,and its steering vector does not satisfy this symmetry relationship.The sound pressure velocity is jointly processed so that the vector array steering vector satisfies this symmetry relationship.On the one hand,the improved MVDR algorithm can be applied to the vector matrix,and on the other hand,based on the combined processing method of the sound pressure and vibration velocity,it can resist the noise in the same direction and reduce the threshold of the SNR ratio of the DOA estimation.Finally,the improved MVDR algorithm was validated by the underwater target detection experiment conducted by Songhua Lake and Qiandao Lake respectively.The echo data of single target and dual target were processed respectively,and the conventional beamforming algorithm,MVDR algorithm and The improved MVDR algorithm processing effect verifies the feasibility and effectiveness of the proposed method in the practical application process.And compared the relationship between the improved MVDR algorithm and the CBF algorithm and MVDR algorithm in the calculation time and efficiency improvement ratio under different spatial angle search step conditions,it is verified that the efficiency improvement is more obvious when the spatial angle search step is smaller in theoretical analysis.