摘要
针对传统的功耗管理方法不能解决实际应用中汇聚节点能量资源有限而易成为网络瓶颈的问题,以及传统排队论的指数分布理论模型在处理无线网络业务流时无法准确描述自相似性问题,提出了一种无线传感器网络汇聚节点动态功耗管理方法.首先理论推导了节点在各个休眠状态的休眠阈值,采用定数截尾样本的一致最小方差无偏估计方法得到汇聚节点在持续空闲时间长度上服从Pareto分布的参数,在不影响网络性能的前提下,利用二维模糊控制器自适应地调整定数截尾样本选用窗口.实验结果表明,该方法在满足网络性能约束的前提下,经动态调整节点的休眠深度,可减少约40%汇聚节点的能量开销、30%的动态功耗管理切换时延,提高约56%的切换成功率,从而避免了汇聚节点易成为网络瓶颈节点的问题,有效延长了汇聚节点的工作寿命.
A novel dynamic power consumption management （DPCM） method is proposed for the sink node of wireless sensor networks （WSN）, in which the sleeping threshold of each node in sleeping status is deduced theoretically by utilizing the uniformly minimum variance unbiased estimation （UMVUE） of fixed truncated samples to obtain the parameters of idle duration that obey the Pareto distribution. On the premise of no effect on performance of WSN, the fixed truncated samples are adaptively adjusted to select the window size by using a two-dimension fuzzy controller. The simulation results show that the proposed DPCM algorithm can dynamically regulate the sleep depth of nodes while satisfying the network performances, reduce about 40% of the energy consumption of the sink node and about 30% switching latency of DPCM, and increase about 56% of the success ratio of switching, thereby the problem that the sink node is easily to become a ＂bottleneck node＂ can be avoided, and the sink node＇s lifetime is effectively improved.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2008年第2期138-142,共5页
Journal of Xi'an Jiaotong University
基金
国家高技术研究发展计划资助项目(2003AA1Z2610)
关键词
无线传感器网络
汇聚节点
动态功耗
节点寿命
wireless sensor networks
sink node
dynamic power consumption
node lifetime
