Energy efficient geographic routing for wireless sensor networks.

A wireless sensor network consists of a large number of low-power nodes equipped with wireless radio. For two nodes not in mutual transmission range, message exchanges need to be relayed through a series of intermediate nodes, which is a process known as multi-hop routing. The design of efficient routing protocols for dynamic network topologies is a crucial for scalable sensor networks. Geographic routing is a recently developed technique that uses locally available position information of nodes to make packet forwarding decisions. This dissertation develops a framework for energy efficient geographic routing. This framework includes a path pruning strategy by exploiting the channel listening capability, an anchor-based routing protocol using anchors to act as relay nodes between source and destination, a geographic multicast algorithm clustering destinations that can share the same next hop, and a lifetime-aware routing algorithm to prolong the lifetime of wireless sensor networks by considering four important factors: PRR (Packet Reception Rate), forwarding history, progress and remaining energy. This dissertation discusses the system design, theoretic analysis, simulation and testbed implementation involved in the aforementioned framework. It is shown that the proposed design significantly improves the routing efficiency in sensor networks over existing geographic routing protocols. The routing methods developed in this dissertation are also applicable to other location-based wireless networks.