Distributed DOA Estimation Based on Manifold Separation Technique in Mobile Wireless Sensor Networks

Abstract

The limited battery capacities of sensor nodes have become the biggest impediment to the applications of mobile wireless sensor networks (MWSNs) over the years. Thus when the sensor nodes of the MWSNs are deployed in the interest region to estimate the direction-of-arrival (DOA), the computational complexity of the proposed algorithm has to be taken into consideration. In this paper, a distributed DOA estimation algorithm using manifold separation technique (MST) and data fusion technique is proposed. MST is a method for modeling the steering vector of antenna arrays of practical interest with arbitrary 2D or 3D geometry. It is the product of a sampling matrix (dependent on the antenna array only) and a Vandermonde structured coefficients vector depending on the DOAs only. In most conditions, the environment of the deployment region of the sensor nodes can be known as a priori, thus the deployment position of the nodes can be determined before deploying the sensor nodes. The topology structure of the MWSNs in each deployment region can be known as a priori, then the sampling matrix can be calculated ``off-line. The information about the sampling matrix can be stored in the system on tap. Then based on the rule of data fusion, the weight value can be formulated as a function of the number of nodes, the snapshot number and the battery capacity of the nodes. Then total cost function can be achieved finally. The Root-MUSIC is used to obtain the DOAs of the incidednt signals. The computer simulation verifies the effectiveness of the proposed algorithm.