Mobile diffusion source tracking in sensor networks

Abstract

Compared to the instantaneous mobile source tracking, the mobile diffusion source tracking is more difficult. In this paper, we give a study on the mobile diffusion source tracking in sensor networks. The CPA realtime localization method, the centroid realtime localization algorithm, the analytic realtime localization algorithm and the tracking method based on PF(Particle Filter) are presented to solve the mobile diffusion source tracking problem. The preconditions, advantages and deficiencies of the methods are given. The performances of different tracking methods are compared in simulations when node densities and sampling intervals are different. The results show that all the proposed methods are valid, while the tracking method based on PF is the most robust method compared to others.