Abstract:
In this work, a novel scheme is designed to exploit the fact that disasters occur rarely, but the sensor network should be up and available for a long time. The proposed scheme is devised for disaster recovery for wireless sensor and actuator networks. The main goal is to minimize battery usage, increase network lifetime, and minimize redeployment. The proposed model works even in the case that the disaster destroys all nodes within the disaster area and also blocks communication unless there is line of sight. The scheme works in an area where the actuators form a regular grid and the sensors are randomly deployed. The scheme tries to maximize network lifetime by reducing idle listening and protocol overhead while there is no disaster and rapidly organizes nodes when the disaster strikes. Operation under disaster includes an alarm phase, a grouping phase, and a routing phase. Grouping mechanism provides data aggregation and reduces the overhead of the actuators. Furthermore, routing mechanism allows nodes along the path to perform duty cycling and prevent them from becoming bottlenecks. The proposed scheme is compared with the well known T-MAC algorithm in terms of network lifetime and shown to produce better results.|Keywords: Wireless sensor networks, Wireless sensor and actuator networks, Power efficient disaster detection, Rare event detection, Idle listening, Duty cycling
