TDOA sensor array optimization using Digital Elevation Model

This paper addresses the problem of optimizing TDOA sensor locations using Digital Elevation Model (DEM). The aim of this work is to obtain an effective procedure to emplace TDOA sensors in a large geographical scale taking the number of sensors, the accuracy of position estimation and the coverage into account. This procedure needs information about the layout and digital terrain model of the area. By using DEMs, two major issues that arise in Cramer-Rao lower bound calculation of a sensor array could be addressed: obtaining accurate locations of source and sensors relative to each other; considering the effect of terrain fluctuations and Earth curvature on obscuring sensor-source visibility. The Orthogonal Learning Particle Swarm Optimization (OLPSO) algorithm is used to find sensor geometries that optimize coverage and accuracy at the same time. Simulations are performed using the DEM model for the City of Mashhad, and the optimized arrays with different number of sensors are determined. The comparison results show that the coverage-accuracy ratio of the optimized sensor arrays is much better than the common splay senor arrays which highlight the importance of using such a sensor placement procedure.