Weighted Centroid Localization Algorithm Based on Maximum Likelihood Estimation

In solving the problem of localizing nodes in a wireless sensor network, we propose a weighted centroid localization algorithm based on maximum likelihood estimation, with the specific goal of solving the problems of big received signal strength indication（RSSI) ranging error and low accuracy of the centroid localization algorithm. Firstly, the maximum likelihood estimation between the estimated distance and the actual distance is calculated as weights. Then, a parameter k is introduced to optimize the weights between the anchor nodes and the unknown nodes in the weight model. Finally, the locations of the unknown nodes are calculated and modified by using the proposed algorithm. The simulation results show that the weighted centroid algorithm based on the maximum likelihood estimation has the features of high localization accuracy and low cost, and has better performance compared with the inverse distance-based algorithm and the inverse RSSI-based algorithm. Hence, the proposed algorithm is more suitable for the indoor localization of large areas.