EKF-Based Adaptive Sensor Scheduling for Target Tracking

For Wireless Sensor Networks (WSN), target tracking is a canonical problem that collaborates signal and information processing to dynamically manage sensor resources and efficiently process distributed sensor measurements. This paper proposes an adaptive sensor scheduling strategy that jointly sets up distribute dynamic clustering, selects the tasking sensor, and determines the sampling interval. The approach utilizes Least-Square (LSQ) in initializing, Extended Kalman Filter (EKF) in tracking accuracy estimation, and adaptive sampling in velocity prediction. Simulation results demonstrate significant improvement in tracking accuracy compared to the non-adaptive approaches.