Weighted Centroid Localization Algorithm Based on Normal Distribution in Wireless Sensor Networks

Abstract

Abstract-Localization algorithms continue to be a challenging and important topic in wireless sensor networks (WSNs). To efficiently exploit the normal distribution for range-free localization, we propose two weighted centroid localization algorithms called Positioning Based on Normal Distribution and Positioning Based on the Network Size. The former uses the normal distribution to reduce weights errors. The latter takes a further step to control the variance in normal distribution to estimate the location. Simulation results show that the proposed weighted centroid algorithms outperform WCL and AMWCL-RSSI (anchor_optimized modified weighted centroid localization algorithm based on RSSI) in terms of the localization accuracy for WSNs with planned deployment of anchor nodes. Experimental results demonstrate that the proposed two algorithms have a small localization error.