Scalable coordination for sensor networks in challenging environments

In sensor networks, the unpredictable dynamics caused by the disruption of communication in harsh environments, the depletion of battery resources, and node mobility make coordination between sensor nodes more challenging than coordination between nodes in typical wired and wireless networks. While the literature is rich in coordination protocols for sensor networks and routing protocols for ad-hoc wireless networks, they assume relatively static environments. Using these protocols in highly dynamic environments leads to increased coordination traffic, high energy-consumption, increased event loss, and excessive latency in reporting events. In this paper we propose SWIFT, a coordination protocol that reacts to changes in nodes' connectivity with a local repair process, significantly reducing the coordination traffic. SWIFT eliminates routing loops through a combination of avoidance and detection mechanisms; and does not require GPS support, which cannot be used in closed environments and consumes significant energy resources. Experimental results reveal that SWIFT outperforms existing protocols and achieves significant energy efficiency, reduces event loss, and the latency in reporting events under various network dynamics.