Direction of arrival estimation and signal recovery for underwater target based on compressed sensing

Direction of arrival estimation(DOA) and signal recovery respectively is the precondition of the underwater target positioning, tracking and recognition. Based on the compressed sensing (CS) theory, a kind of method for DOA estimation and signal recovery is proposed using the multiple snapshots processing of the array data in frequency domain. Firstly, the receiving array data are transformed to frequency domain, and the selected multiple snapshots frequency domain data are regarded as the measurements of the CS. Then according to the frequency, searching bearing and array manifold, the overcomplete array manifold matrix is constructed as the sensing matrix of the CS. Finally, both the target signal and power of the searching bearing are estimated by the convex optimization algorithm to complete DOA estimation and signal recovery. Wideband simulation results show that the proposed method has a number of advantages over the minimum variance distortionless response (MVDR) method under the same conditions, including improved robustness to noise, limitations in sensors and snapshots number. And the correlation coefficient of the estimated signal and spectrum is higher than that of the MVDR method. Experiment results in real environments verify that the proposed method performs more effectively in the detection of faint targets than the MVDR method and can be used in real sonar system.