Robust Mobile Location Estimator with NLOS Mitigation using Reconstruction and Regression

In recent years, the rapid development of microelectronics technology, computer science and technology, wireless communication technology, promote the progress of low-cost, low-power wireless sensor network (WSN)technology. Its application in various fields, especially in target tracking, has attracted more and more attention of scholars. Non-line-of-sight (NLOS) propagation is a valuable question in wireless sensor network positioning. Especially in indoor environments where positioning accuracy is highly demanded, precise positioning of moving targets is still a difficult problem due to the continuous switching of LOS and NLOS environments. Commen filters such as the extended Kalman filter (EKF) has a poor performance because they are sensitive to outliers. In these paper, we propose a robust mobile terminal location method based on reconstruction and Regression. We identify the propagation channal with GMM. Then a limiting filter is employed to migate the NLOS error, and regression is uesed to smoothing the positioning result. The simulation results show that the method has good positioning accuracy.