Least Square Localization Method Based on Anchor Nodes Optimization Selection

During the process of Least Square localization, some negative factors may give rise to different levels of noise, such as the environmental noise, the reflection, refraction, multipath and non-line-of sight (NLOS) complex propa gation of wireless signal, and the limitation of distance estimation method. And they also lead to low localization accuracy of Least Square localization. For this problem, this paper proposes an improved Least Square localization method, which is called Least Square localization based on anchor nodes optimization selection through minimum standard deviation (LS-ANOS). In LS-ANOS method, nanoLOC-based Symmetric Double Sided Two Way Ranging (SDS-TWR) is utilized to conduct distance estimation repeatedly between unknown nodes and anchor nodes. And statistical computation is performed on these distance estimation results. Then, from the influential mechenism of input measurement noise on localization result, the paper adopts slide window-based single scanning strategy to optimize the selection of the distance estimation result with higher quality and the corresponding anchor nodes. Lastly, based on the least square localization computation, it gets the accurate localization result. Simulation and experimental results demonstrate that the proposed method could improve the accuracy of Least Square localization method effectively.