An Improved NLOS Identification and Mitigation Approach for Target Tracking in Wireless Sensor Networks

Abstract

Target tracking has wide-ranging applications in fields using wireless sensor networks. However, localization accuracy is adversely affected by the non-line-of-sight (NLOS) effect. Thus, we propose a three-step localization approach to target tracking to identify and mitigate the NLOS effect. A Bayesian sequential test is designed to identify whether the measurement data are affected by this effect. On the basis of the identified measurement condition, we smooth the measurement range and mitigate the NLOS effect using a modified Kalman filter (MKF). After adjusting the measurement noise covariance and prediction covariance by using an established measurement equation, we apply the MKF, which is a standard Kalman filter with updated parameters. After the distances between the target and the sensor nodes are estimated by the MKF, the final estimated target position can be obtained using a residual weighting algorithm. Experimental and simulation results show that the proposed approach is superior to other methods that do not identify the propagation condition, and it can effectively improve the localization accuracy.