A Maximal Concurrency and Low Latency Distributed Scheduling Protocol for Wireless Sensor Networks

Abstract

Existing work that schedules concurrent transmissions without collisions suffers from low channel utilization. We propose the Optimal Node Activation Multiple Access (ONAMA) protocol to achieve maximal channel spatial reuse through a distributed maximal independent set (DMIS) algorithm. To overcome DMIS's excessive delay in finding a maximal independent set, we devise a novel technique called pipelined precomputation that decouples DMIS from data transmission. We implement ONAMA on resource-constrained TelosB motes using TinyOS. Extensive measurements on two testbeds independently attest to ONAMA's superb performance compared to existing work: improving concurrency, throughput, and delay by a factor of 3.7, 3.0, and 5.3, respectively, while still maintaining reliability.