An Improved DOA Estimation Method Based on Compressed Sensing for the Distributed Array

Direction of Arrival (DOA) estimation is important for target detection in the distributed array, however, traditional DOA algorithms based on subspace, such as Multiple Signal Classification algorithm (MUSIC) and Estimating Signal Parameter via Rotational Invariance Techniques (ESPRIT), cannot be applied directly to distributed array since the emergence of unwanted grating lobes will make the estimation impossible. Orthogonal Matching Pursuit (OMP) algorithm is a kind of Compressed Sensing (CS) algorithm which utilized the sparsity of the echo signals, it could be applied for DOA estimation of the distributed array. Nevertheless, the accuracy of the reconstructed sparse matrix will decrease when the Signal-Noise-Ratio (SNR) is low or the number of snapshots is small, which could be improved by setting a smaller residual value to achieve more iterations. In this paper, an improved recovery method based on Particle Swarm Optimization (PSO) algorithm is proposed for the distributed array. This method could calculate the iteration termination condition threshold by PSO algorithm and avoid the deteriorate of accuracy and increase of complexity caused by excessive iteration. Simulation results showed that this algorithm could provide an accurate reconstructed matrix for low SNR and snapshots while avoiding the computation complexity.