Multi-sensor node fusion localization using unscented kalman filter in rough environments

Localization information of wireless sensor network is an important application in indoor complicate environment. In line-of-sight (LOS) environment, localization accuracy is very high. However, the measurement may be contaminated by nonline-of-sight (NLOS) propagation, which will result in localization accuracy degrades. To cover come this problem, this paper proposes a modified Kalman filter(MKF) localization method based on Unscented Kalman filter(UKF). Firstly, test statistic method is employed to identify the NLOS condition. Then we smooth the measurement range using linear Kalman filter (LKF) and mitigate the NLOS effect using MKF. Finally, the method of UKF is used to determine the localization of unknown motion node. The validity of the proposed approach is demonstrated through the numerical simulation and experiment.