Covariance Difference Matrix-based Sparse Bayesian learning for Off-grid DOA Estimation with Colored Noise
Meihong Pan, Gong Zhang, Zhentao Hu · 2019 IEEE MTT-S International Microwave Biomedical Conference (IMBioC) · 2019
Most of the existing direction of arrival (DOA) estimation methods are applicable in the presence of Gaussian white noise situation. The precision of the traditional sparse representation based DOA estimation methods are limited to the grid interval. To address these problems, the covariance differencing method is applied and the colored noise component is eliminated by forming the difference of the original and the transformed covariance matrices of the observation data. And then an off-grid DOA estimation algorithm based on the eigenvectors corresponding to the positive eigenvalues of covariance difference matrix is proposed. The sparse Bayesian learning strategy is adopted to solve the off-grid DOA estimation model. The proposed method can suppress the colored noise and avoid the pseudo peaks effectively. Simulation results demonstrate the improvement of estimation accuracy.