A High Resolution DOA Estimation Based on Random Array in Underwater Multipath Environment

Abstract

In the underwater sonar signal processing technology, linear sonar array should conform to Nyquist sampling theorem, otherwise the problem of azimuth ambiguity will occur in the underwater Direction of Arrival (DOA) estimation. To solve this problem, a kind of nonuniform linear array random spacing array is studied, and the application of this array structure in DOA estimation of underwater targets is studied. In addition, after different transmitted signals of underwater active sonar detection target are scattered and refracted through underwater acoustic channel, the echo signal will appear coherent, which seriously affects the detection accuracy of underwater target. To solve this problem, this paper proposes Compressed Sensing Basic Pursuit De-Noising (CS-BPDN) algorithm based on random array. In this paper, CS-BPDN algorithm and Multiple signal classification (MUSIC) algorithm are combined with random spacing array to estimate the DOA of underwater targets by using the acoustic ray propagation simulation in BELLHOP model. The simulation results show that the CS-BPDN algorithm based on random spacing array can effectively deal with the problem of mixed coherent signals caused by underwater multipath effects.