Research on Node Localization in Wireless Sensor Networks Based on Kalman Filter

This paper Aims at reducing the impact of the ranging error and Random error of the node localization nodes and additional error accumulated caused by localization algorithm on the node localization accuracy in wireless sensor network. Firstly this paper uses the minimum cumulative related distance error in LSL (LSL-MCR) algorithm to improve the node localization precise, and than this paper uses a hybrid algorithm which combines least squares estimation with Extended-Kalman Filter, finally the two different algorithms are compared by simulation, the simulation result shows that Kalman filter can further improve the node localization accuracy, and it is especially fit for the cases with network node density and the low percentage of beacon nodes.