A Modified RSSI-Based Indoor Localization Method in Wireless Sensor Network

Node location is one of the basic problems in wireless sensor networks (WSN). We investigate the wireless localization methods based on providing received signal strength index (RSSI) measurements between a mobile node and several access nodes in a WSN. In order to estimate the distance between two nodes, a propagation model is used to transform RSSI into the corresponding distance that the mobile node is away from the access points according to the RSSI signals. Next, the mobile node coordinate can be obtained by any location algorithm such as weighted B-box considered as the better method compared with the trilateraion method. To achieve a better location result, Gaussian filtering, piecewise fitting, and total least square estimation are introduced to improve precision of estimated model parameters, and iterative algorithm is adopted to further reduce positioning errors. Several simulation results reveal that the MSE of the proposed method is cut off at lease 2/3 while comparing to the B-box algorithm.