Sparse recovery using an iterative Variational Bayes algorithm and application to AoA estimation
Ahmad Bazzi, Dirk T. M. Slock, Lisa Meilhac · 2016
This paper presents an iterative Variational Bayes (VB) algorithm that allows sparse recovery of the desired transmitted vector. The VB algorithm is derived based on the latent variables introduced in the Bayesian model in hand. The proposed algorithm is applied to the Angle-of-Arrival (AoA) estimation problem and simulations demonstrate the potential of the proposed VB algorithm when compared to existing sparse recovery and compressed sensing algorithms, especially in the case of closely spaced sources. Furthermore, the proposed algorithm does not require prior knowledge of the number of sources and operates with only one snapshot.