Off-grid direction of arrival estimation based on weighted sparse Bayesian learning

In this paper, a weighted sparse Bayesian learning algorithm for off-grid direction of arrival (DOA) estimation is proposed. By utilizing the relationship between the noise subspace and the overcomplete basis matrix, the weights are designed and treated as the hyperprior knowledge of the signals, which changes the variance of the Laplace distribution of the signal, i.e., the average power of the signal, to enhance sparsity of the solution and improve the estimation accuracy. Compared with the original off-grid sparse Bayesian method, the proposed one can not only improve the performance, but also give a faster DOA estimation. Simulation results demonstrate the efficiency of the proposed method.