Intelligent WSN Localization using Multi-Linear Regression and a Mobile Anchor Node

As the applications of Wireless Sensor Network are increasingly associated with the rapidly growing field of the Internet of Things, resolving the problem of precisely locating sensor nodes has become a pressing need. As a result, several location identification methodologies have been proposed to address this need. This paper proposes a machine learning-based algorithm that allows the sensor node to identify its own location accurately using a single mobile anchor node. The mobile anchor node moves along a relatively short straight path, reducing the energy consumption associated with its movement as well as the total time needed to identify the locations of all sensor nodes. The proposed machine learning model is based on the Multi-Linear Regressing model and it is trained offline, reducing the burden on the resource-constraint sensor nodes. The performance evaluation results show that the proposed algorithm can achieve location identification accuracy of less than half a meter.