DISTRIBUTED KALMAN FILTER-BASED TARGET TRACKING IN WIRELESS SENSOR NETWORKS

The tracking of mobile targets using Distributed Kalman Filters in a Wireless Sensor Network (WSN) is addressed in this paper. In contrast to the Kalman Filter implementations reported in the literature, our approach has the Kalman Filter running on only one network node at any given time. The knowledge learned by this node, i.e. the system state and the covariance matrix, is passed on to the subsequent node running the filter. Since a finite subset of the sensor nodes is active at any given time, target tracking can be accomplished using lower power compared to centralized implementations of the Kalman Filter. Numerical simulations demonstrate that the proposed algorithm is robust to measurement noise and changes in the velocity of the target. The results in this paper show that the proposed technique for target tracking will result in significant savings in power consumption and will extend the useful life of the WSN.