An Energy Efficient Surveillance System for Wireless Sensor Networks

The miniaturization of sensor devices, combined with their ability to gather information about the physical environment in which they are deployed, makes sensor nodes wellsuited for unmanned reconnaissance missions. However, the software support required to use a wireless sensor network for this purpose is still lacking. In this paper, we describe a complete framework we have designed that provides the software support necessary for using wireless sensor networks for reconnaissance. The primary goal of this framework is to enable a sensor network to track objects in an energy-efficient and stealthy manner, with minimal false alarms. The framework minimizes false alarms by aggregating the reports from a group of neighboring sensor nodes. It achieves energy-efficiency and stealthiness by placing the sensor nodes in a low power consuming state in the absence of external events, and by minimizing the number of messages transmitted in that state. We have evaluated our implementation by deploying 70 nodes running our software in an open field. Our results show that for the configuration we used, the number of false alarms reduces to zero when the degree of in-network aggregation is 3. Our power management strategy is capable of extending the lifetime of the sensor nodes by up to 900% when the nodes operate for an hour per day. The number of messages in the idle state is nearly zero, confirming the stealthiness of our power man-