A Message Passing Algorithm for Off-grid DOA Estimation Based on UWB System

The Ultra Wideband (UWB) technology has attracted extensive attention and is of great significance for positioning. Direction of arrival (DOA) estimation can improve positioning accuracy of UWB. At present, most researches on DOA estimation algorithms are based on sparse Bayesian learning (SBL) framework. This paper mainly proposes a solution for off-grid DOA estimate of UWB system. We introduce Dirichlet process (DP) prior to cluster the source UWB signals. To describe the problem, we design a factor graph. Mean field (MF) rule is used for the factor graph and a message passing algorithm is obtained. Experimental results prove that the algorithm proposed in this paper is superior.