A Robust Wireless Sensor Network Localization Algorithm in NLOS Environment

Reliable location estimation is an essential and fundamental issue for wireless sensor network. In order to find a localization algorithm robust to non-line-of-sight (NLOS) propagation errors in sensor network system, we analyzed some latest localization algorithms. Considering the application environments of the network, the energy consumption of nodes, the amount of computation and the location accuracy, we present a novel localization algorithm with the method of select residual weighting (SRwgh). SRwgh algorithm can reject the range measurements with large errors to improve the localization performance without identifying the NLOS error(s) or knowing channel characters. When the number of range measurements is larger than 4, this algorithm needs less amount of computation than that of conventional residual weighting algorithm and gets almost the same performance. The algorithm is compared with three other generic algorithms. The performance of the algorithms is test by simulation in both indoor and outdoor environments. Simulation results show that it is robust to the NLOS error(s) and provides quite well location accuracy.