Collaborative Re-Localization Method in Mobile Wireless Sensor Network Based on Markov Decision Process

Localization in Mobile Wireless Sensor Networks (WSNs), particularly in areas like surveillance applications, necessitates triggering re-localization in different time periods in order to maintain accurate positioning. Further, the re-localization process should be designed for time and energy efficiency in these resource constrained networks. In this paper, an energy and time efficient algorithm is proposed to determine the optimum number of localized nodes that collaborate in the re-localization process. Four different movement methods (Random Waypoint Pattern, Modified Random Waypoint pattern, Brownian motion and Levy walk) are applied to model node movement. In order to perform re-localization, a server/head/anchor node activates the optimal number of localized nodes in each island/cluster. A Markov Decision Process (MDP) based algorithm is proposed to find the optimal policy to select those nodes in better condition to cooperate in the re-localization process. The simulation shows that the proposed MDP algorith