Particle Filtering Based on Sign of Innovation for Tracking a Jump Markovian Motion in a Binary WSN

This paper deals with target tracking in a binary wireless sensor network. In this contribution, the target motion is represented by a non linear model based on a jump-Markovian direction. The observation of the trajectory based on the target signal strength is performed by binary sensors. In particular, only one sensor emits one bit by instant what increases the longevity of the wireless sensor network. A particle filter based on the sign of innovation is proposed to track the target motion. Relying on exhaustive simulation results, the proposed SOI-PF exhibits very attractive transient and steady state performance. In deed, for all the considered tracking contexts a high convergence rate and an accurate estimation of the target position are obtained.