Target tracking in wireless sensor networks using sequential implementation of the extended Kalman filter

Abstract

The centralized extended Kalman filter is a commonly used approach for target tracking in wireless sensor networks, which usually consumes heavy computing energy on the leader of the tracking cluster. In this paper, we present a target tracking approach using wireless sensor networks based on sequential implementation of the extended Kalman filter. At every tracking time, each member of the tracking cluster transmits its measurement to the leader. The leader utilizes extended Kalman filter to update the current target state estimate whenever it receives a measurement, and completes the current tracking process until it receives the last sensor measurement. Simulations results demonstrates that the proposed target tracking approach can achieve more accurate tracking accuracy than the centralized extended Kalman filter-based tracking approach, but reduce computation time.