Improved Subspace Direction-of-Arrival Estimation in Unknown Nonuniform Noise Fields

This paper improves the classic subspace DOA estimation methods to combat unknown nonuniform noise. Utilizing an approximate orthogonality between the signal subspace and a tailored eigen-space of the array covariance matrix in high signal-to-noise ratio (SNR) conditions, we modify the classic multiple signal classification (MUSIC) and root-MUSIC algorithms to be competent in unknown nonuniform noise. Compared to the MUSIC and root-MUSIC methods, the proposed methods are able to achieve significant better performance in unknown nonuniform noise environments. Simulation results show that the two proposed methods significantly outperform the MUSIC and root-MUSIC methods in the whole SNR range and approach the Cramer-Rao bound (CRB) at high SNR.