Scoped Bellman-Ford Geographic Routing for Large Dynamic Wireless Sensor Networks

Routing is a fundamentalproblem in wireless sensor networks. Most previous routing protocols arechallenged when used in large dynamic networks as they suffer fromeither poor scalability or the void problem. In this paper, we propose anew geographic routing protocol, SBFR (Scoped Bellman-Ford Routing), forlarge dynamic wireless sensor networks. The basic idea is that each nodekeeps a view scope of the network by computing distance vectors usingthe distributed Bellman-Ford method, and maintains a cost for routing tothe sink. When forwarding a packet, a node picks the node with minimumcost in its routing table as a temporary landmark. While achieving goodscalability, it also solves the void problem in an efficient mannerthrough the combination of Bellman-Ford routing and cost-basedgeographic routing. Analytical and simulation results show that SBFRoutperforms other routing protocols not only because of its robustnessand scalability but also its practicality and simplicity.