DOA Estimation of Quasi-Stationary Signals Using Sparse Signal Reconstruction

Abstract

Quasi-stationary signals have been widely used in practical applications, for example, microphone array signal processing, which have time-varying statistical property while staying static within a short period of time. In this paper, a sparse signal reconstruction algorithm based on Khatri-Rao (KR) product theory is proposed for direction of arrival (DOA) estimation of quasi-stationary signals. The algorithm transforms the covariance matrix of the received data by KR product, and DOA estimation is processed as a virtual array multiple measurement vector model after dimensionality reduction, denoising and realization. Redundant dictionary is constructed by the sparseness of sources in spatial, and the orthogonality of subspace is used to determine weights, which enhance the sparsity of solution vectors. Finally, DOA estimation is solved by the inverse problem of sparse signal reconstruction. Simulation results verify the effectiveness of the proposed algorithm, which can be used for DOA estimation in undetermined cases and has a small estimation error in the condition of low signal-to-noise ratio.