An Energy-Efficient Self-Adaptive Duty Cycle MAC Protocol for Traffic-Dynamic Wireless Sensor Networks

摘要

A plenty of Medium Access Control (MAC) protocols deal with static traffics with low traffic load. The performance of these protocols drops significantly when network traffic become dynamic or in high traffic load. In this paper, we propose a new MAC protocol called Self-Adaptive Duty Cycle MAC (SEA-MAC) by introducing dual adaptive mechanisms: (1) An Adaptive Scheduling (AS) mechanism which makes the nodes' active duration adaptive to variable traffic load, thus enabling SEA-MAC to resiliently schedule data transmission for the Sleep period. The algorithm is designated to schedule more data transmission in bursty and high traffic load, thus enabling rapid dissemination of data and reduction of latency. While under the light traffic load, nodes enter the Sleep mode timely, mitigating idle listening and saving energy. (2) A Self-Adaptive duty cycle mechanism which further adjusts the duty cycle and makes SEA-MAC adaptive to the dynamic traffic loads. When network experience with very low or very high traffic load, the protocol further adjusts duty cycle dynamically based on the traffic load, thus reducing inefficient duty cycle or reduce end-to-end delay. Experiment results show that the SEA-MAC is a great advancement compared with AS-MAC and RI-MAC protocols on the performance for unicast scenarios, especially under heavy unicast traffic load. SEA-MAC can reduce average and max end-to-end delay up to 50.90 and 68.20% respectively, lower energy consumption up to about 16.28%, and increase PDR up to about 16% compared with AS-MAC. SEA-MAC superiors to RI-MAC in average end-to-end delay by up to 80% and in maximum end-to-end delay up to 90% for all the traffic load in question. SEA-MAC outperforms RI-MAC significantly in average duty cycle under medium and heavy traffic load.