A Sparse Recovery Method of DOA Estimation for Uncorrelated Sources with Multi-snapshot

Abstract

In this paper, a novel sparse recovery method is proposed for direction of arrival (DOA) estimation on the assumption that source signals are uncorrelated and the number of snapshots is enough. We demonstrate in this paper how to solve DOA estimation problem with multiple measurement vectors (MMV) from the perspective of the correlation rather than from the view of data which appears in almost all sparse signal recovery methods. By using this method we remove the effect of noise and attain both DOA estimation and signal power. Theoretical analysis and experimental results demonstrated that our approach has a number of advantages compared other source localization schemes, including increased resolution, lower bias and variance, unknowing the number of sources, better performance in low SNR, decreased computational complexity.