点击次数：

论文类型：期刊论文

发表时间：2012-10-01

发表刊物：TRANSACTIONS ON EMERGING TELECOMMUNICATIONS TECHNOLOGIES

收录刊物：SCIE、EI、Scopus

卷号：23

期号：6

页面范围：585-597

ISSN号：2161-3915

关键字：backoff; proportional fairness; shadow price; wireless sensor networks

摘要：The many-to-one and hop-by-hop traffic pattern in wireless sensor networks exhibits a unique funnelling effect that results in a significant increase in transit traffic intensity, package collision among different links and congestion as closing toward the sink. Some techniques, such as congestion control and aggregation, can help counter these problems. However, they cannot fully alleviate funnelling effect. In this paper, we use a different and novel approach to solve this problem. The proportional fairness backoff scheme is based on the carrier sense multiple access/collision avoidance protocol and makes the network's throughput maximum with proportional fairness (data transmission fairness between different nodes). Furthermore, this proportional fairness backoff scheduling scheme can dynamically adjust each sensor node's backoff window on the basis of Kelly's shadow price theory. Further, on the basis of the newly designed throughput analysis model, we discuss the throughput of the carrier sense multiple access/collision avoidance-type network. With theoretical analysis and extensive simulations, we show that proportional fairness backoff can effectively mitigate funnelling effect and then improve the network's throughput; at the same time, it keeps effective throughput increase even if the network is high-loaded. Copyright (c) 2012 John Wiley & Sons, Ltd.