Abstract:

Direction of arrival (DOA) cannot be estimated accurately by multiple signal classification (MUSIC) algorithm under the condition of less snapshot numbers or the presence of coherent sources, while with the increase of snapshot numbers or the decrease of signal-to-noise ratio, the resolution performance of the compressive sensing (CS) method is not stable enough and the rate of accuracy is limited. A method based on CS and MUSIC is proposed and the abbreviation is called CS-MUSIC. The link between CS and MUSIC for different snapshot numbers is established in the proposed method which constructs a new orthogonal space and achieves sharp peaks. Theoretical analysis and simulation results show that the proposed method has estimation accuracy to overcome the defects of the traditional methods under the condition of different snapshots and it is robust to noise.