An Efficient Monte Carlo-Based Localization Algorithm for Mobile Wireless Sensor Networks

The aim of this paper is to propose a localization algorithm in which nodes are able to estimate their speeds, directions and motion types. By this way, node's next state can be estimated and the particles can be distributed closer to the predicted locations. Hence, accuracy and the precision of the localization are increased considerably. Sequential Monte Carlo method is used to represent the posterior distribution of a node's possible locations with a set of weighted samples. The developed algorithm is designed for a general network where no restrictions are made on the densities and the distributions of the nodes. Eight localization algorithms, namely Centroid, APIT, Amorphous, DV-Hop, MCL, SMCL, MCB and the developed algorithm SMCLA are implemented and compared for different parameters and mobility models. Simulation results show that the developed algorithm performs best for each of the simulation comparisons.