Parameter estimation of incoherently distributed source based on block sparse Bayesian learning

In practical array signal processing applications, the performance of DOA (direction-of-arrival) estimation methods is known to degrade severely in the presence of angular spread. In this paper, a new approach of estimating parameter via block sparse Bayesian learning is proposed for multiple incoherently distributed sources. Unlike traditional subspace based methods, the new technique makes use of a sparse representation of the received data with a perturbed overcomplete dictionary. Specifically, after using the temporal correlation between snapshots, the central DOA is estimated by using a Bayesian learning algorithm. The new method is able to mitigate the influence of angular spread, and its performance is demonstrated from numerical simulations.