Energy management in sensor networks for continuous monitoring applications

Wireless sensor networks provide a versatile computing platform for applications such as environmental monitoring or military field surveillance. These networks are typically deployed in remote, hostile, or hazardous environments where sensor nodes may not have access to power sources. Hence, energy efficiency is the most critical issue in wireless sensor networks. In this thesis, we study energy management schemes in sensor networks for continuous monitoring applications, namely, applications employing continuous sensing and data delivery, where sensors periodically send their data to a base station. We study sleep/wake scheduling for low-duty cycle sensor networks. Prior work on sleep/wake scheduling has either assumed that an underlying synchronization protocol can provide perfect synchronization, or assumed an upper bound on the clock disagreement, and used it as a guard time . We show that synchronization error is non-negligible, and using a conservative guard time is energy wasteful. We then study the optimal sleep/wake scheduling scheme with consideration of the synchronization error. We first investigate the sleep/wake scheduling problem for single hop communication scenario, then extend our study to multihop communications. We also study data gathering in sensor networks. For continuous monitoring applications, a tree-based topology is often used to collect data from sensor nodes. We study the construction of a data gathering tree to maximize the network lifetime, which is defined as the time until the first node depletes its energy. The problem is shown to be NP-Complete. We then give an approximation algorithm, which terminates in polynomial time and is provably near optimal