Abstract

Target localization is one of several applications that capitalizes on the sensor network framework. There have been a number of investigations for target localization and tracking in sensor network context. These ideas, however, were mostly experimented only in centralized processing scenarios. The study of the parameters that affect both the localization accuracy and the energy expenditure is significant for the decentralized system design. In this paper, we focus on Range Difference (RD) based target localization due to the feasibility of realtime performance. Our study will emphasize the analysis of the localization performance associated with decentralized processing. Assuming a densely and uniformly distributed sensor placement, we derive the Mean Square Error when the cluster is activated. The influential parameters which are the number of participating sensors, the distance between reference sensor and the actual target location, and the size of the activated clusters are explicitly presented in the expression. The derived expression is verified by simulation results. We also discuss the benefit earned from the analysis and suggest a solution for an optimization problem that would aid the large scale system design.

Date

January 1, 1970

Authors

Chartchai Meesookho, Urbashi Mitra, Shrikanth Narayanan

Journal

Simulation

Volume

1

Pages

0.8