Linear-regression-based Weighted Centroid Localization Algorithm in Wireless Sensor Network*

摘要

Based on establishment of mathematical models,this paper proposed a new localization algorithm based on range-free mechanism which called linear-regression-based weighted centroid localization algorithm.This algorithm improved weighted centroid localization algorithm by the use of hops between nodes, centralizaion and then calibrated the nodes' position by the use of linear regression. Finally, the algorithm corrected the position through certain parameters. On matlab simulation platform, comparing with the original centroid localization algorithm, this new algorithm's average localization error dropped by more than half and localization ratio was increased to close to 1.