Distributed DOA Estimation in Wireless Sensor Networks Using Randomized Gossip Method

Abstract

In this paper, we consider the problem of direction of arrival (DOA) estimation achieved by a distributed way in wireless sensors networks. The goal for each node is to detect targets based on its local information and that of its neighbors through some iteration. Classic estimation methods, such as maximum likelihood (ML) algorithm, are not suitable here because there is often a requirement of a central unit to obtain the optimal solution. We propose a new distributed DOA estimation algorithms based on the randomized Gossip method, the goal of which is to realize the conventional Capon method by a distributed way. The proposed algorithm does not require any constraint on the network geometries, thereby making it suitable for distributed signal processing in large wireless sensor networks. The given simulation results illustrate the main characteristics of the proposed algorithm, including DOA resolution and mean square error (MSE) performance.