Sleep-based Topology Control for Wakeup Scheduling in Wireless Sensor Networks

Wireless sensor network applications require both energy-efficiency and low latency for reporting urgent but rare events. Various wakeup scheduling schemes have been proposed to save energy by employing duty cycles and to reduce end-to-end delay by synchronizing nodes\´ wakeup times and shortening setup latencies. Most sensor applications incur a predominant "convergecast" traffic pattern in which several close-by sensor nodes simultaneously send reports to the base station, which generates "spatially-correlated contention." This contention, specific to sensor networks, can significantly prolong delay, degrade throughput, and impair energy-efficiency. However, most wakeup scheduling and MAC protocols proposed for sensor networks did not address this contention and can be adversely affected. We propose a distributed topology control technique to schedule nodes\´ wakeup time slots, and design a MAC protocol to benefit from this topology control for improving energy-efficiency and delay, and efficiently handling spatially-correlated contention. Through analysis and simulations, we show that our topology control and MAC can either reduce the end- to-end delay by roughly half or extend the network lifetime by three times, when compared against the nearest competing approach, and achieves twice the throughput under spatially-correlated contention.