Sift: A MAC Protocol for Event-Driven Wireless Sensor Networks

Abstract. Nodes in sensor networks often encounter spatially-correlated contention, where multiple nodes in the same neighborhood all sense an event they need to transmit information about. Furthermore, in many sensor network applications, it is sufficient if a subset of the nodes that observe the same event report it. We show that traditional carrier-sense multiple access (CSMA) protocols for sensor networks do not handle the first constraint adequately, and do not take advantage of the second property, leading to degraded latency as the network scales in size. We present Sift, a medium access control (MAC) protocol for wireless sensor networks designed with the above observations in mind. We show using simulations that as the size of the sensor network scales up to 500 nodes, Sift can offer up to a 7-fold latency reduction compared to other protocols, while maintaining competitive throughput.