Multi-hop-based Monte Carlo Localization for Mobile Sensor Networks

Many low-cost localization techniques have been proposed for wireless sensor networks. However, few consider the mobility of networked sensors. In this paper, we propose an effective and practical localization technique especially designed for mobile sensor networks. Our system is based on the sequential Monte Carlo method, but dissimilar to other conventional localization schemes, our algorithm covers a large sensor field with very few anchor nodes by information flooding. The algorithm works without knowledge of the maximum transmission range, and covers some of the problems caused by the flooding beacons. We discuss factors to implement the algorithm in the real world and present several solutions. Our mechanism is implemented in a real environment, and its feasibility is validated by experiments. The simulation results show that our algorithm outperforms conventional Monte Carlo localization schemes by decreasing estimation errors by up to 50%, and the overhead of the algorithm could be minimized by appropriately adjusting the system parameters.