CMSA: A CENTRALIZED MAINTENANCE STRATEGY WITH ANTICIPATION FOR WIRELESS SENSOR NETWORKS

Providing a continuous service is the main requirement for many application types deployed in Wireless Sensor Networks (WSNs). In this paper, we study two maintenance strategies using a small number of mobile maintainer robots to restore the WSN coverage and connectivity upon a sensor failure: the Centralized Maintenance Strategy (CMS) and the Centralized Maintenance Strategy with Anticipation (CMSA). The CMS and CMSA are based on the Centralized Manager Algorithm [1] used to detect, report sensor failures and coordinate the movement of robots. In CMSA, the predictive version of CMS, a selected robot is chosen as a manager to anticipate the sensor failures and schedule the available maintainer robots to repair them before they happen. To predict the lifetime of a sensor node, we propose an energy state model that represents the behavior of a sensor node based on Markov Chain. We use this model, validated by simulations, to predict the energy consumption by a sensor node and consequently the lifetime of a wireless sensor node. The simulation results show that the CMSA ensures a null dysfunction network time and a message overhead lower than the classical centralized manager strategy.