Using ℓ1, 2 mixed-norm MUSIC based on compressive sampling for direction of arrival estimation

High-resolution direction of arrival (DOA) estimation has always been an issue in signal processing field. Most of conventional methods such as Multiple Signal Classification (MUSIC), suffer from lack of resolvability in low signal to noise ratio (SNR) or small number of snapshots whereas computation cost of new methods are considerable. The Compressive MUSIC (CS-MUSIC) algorithm deals with joint sparse problems with employing both array processing and compressive sensing (CS) recovery algorithms to achieve a high-resolution result. In this paper a hybrid method is used for DOA estimation with employing ℓ1, 2 mixed-norm minimization jointly with MUSIC. The performance of the proposed method in terms of resolution and resolvability for both coherent and non-coherent sources is compared with conventional MUSIC and CS-MUSIC. ℓ1,2-MUSIC overcomes the coherency barrier and has high robustness to noise for few numbers of snapshots.