Wireless sensor networks, which consist of a large number of sensor nodes communicating with each other in order to sense the environment, is an emerging field in the area of wireless networking. One of the significant objectives in the design of these wireless networks is the efficiency of energy consumption. Since these networks are densely deployed, sleep scheduling, which puts some sensor nodes into sleep mode and uses only a necessary set of active nodes, is a commonly used technique to extend the network lifetime. A sleep scheduling algorithm should be distributed, simple, scalable and energy efficient. In this thesis, we investigate the problem of designing a sleep scheduling algorithm which extends the network lifetime while maintaining a user defined coverage and connectivity. We consider three basic units of a sensor node (sensing, processing, and communication) independently and turn a unit on whenever it is necessary. Furthermore, we analysed the expected overlap of a node's sensing area by its neighbors' sensing areas and created a predictive algorithm in the decision of a node's status. By this way, we reduced the number of messages for maintenance procedures. We evaluated our algorithm via simulations and compared the performance of our algorithm with another distributed algorithm which maintains both coverage and connectivity. As a result, we observed that our algorithm provides a significant increase in the network lifetime and overcomes the other algorithm in all cases.

DATE: 18 July, 2007, Wednesday@ 10:00
PLACE: EA 409