ABSTRACT Tracking of a target is exigent and well-investigated application of wireless sensor networks due to diversified applications. Various applications demand proper estimation of a change in target‘s position or precise distance measurements from each sensing nodes to the target. Such essentials are necessary to be measured and must be sent to the base station for subsequent processing. Accurate target tracking is constrained due to limited resources in a sensor network, aging, faults in sensors, environmental as well as process noises, etc. Moving target detection and tracking requires coordination among nodes in order to achieve high tracking efficiency. A multi-step tracking model of the Kalman filter (KF) and particle swarm optimization (PSO) is proposed which estimates the trajectory of the target and provides close position tracking of the target. Different simulation paths and combinations are considered to investigate the efficacy of the proposed approach. It is observed from the results that the proposed multi-step KF-PSO tracking model outperforms the standard KF by significantly reducing the root-mean-square error and thus improves 6–26% of tracking efficiency for different target trajectories.
|