Sensor Placement For Wireless Localization

Abstract

Given a number of sensors, their characteristics, how should the sensors be deployed so as to minimize the location estimation errors. Yet, the design of sensor deployment techniques still receives little attention. In this paper, we address the problem of optimal deployment of sensors using tight frames for the minimization of the mean square error on the user location. By employing, some of the properties of tight frames, it is seen that the existence of an optimal geometry is governed by the fundamental parseval equality of tight frames. A CRLB-based algorithm utilizing the properties of tight frames and a coordinate descent method is presented for placing sensors which minimize the mean square error of multiple users in a coverage area. Finally, simulation analysis is performed to illustrate the achievable gains in minimizing the mean square error of users by optimally placing sensors using tight frames.