采用任意播方式的机会数据聚集算法  

Opportunistic Data Aggregation Algorithm Using Any-cast

作　　者：王培[1,2] 肖明军[1,2] 黄刘生[1,2] 

机构地区：[1]中国科学技术大学计算机科学与技术系,安徽合肥230027 [2]中国科学技术大学苏州研究院,江苏苏州215123

出　　处：《小型微型计算机系统》2010年第9期1712-1716,共5页Journal of Chinese Computer Systems

基　　金：国家"九七三"重点基础研究发展规划项目(2006CB303006)资助;中科院知识创新工程项目资助;中国博士后基金项目(20080430776)资助;国家博士学科点专项科研基金项目(20070358075)资助

摘　　要：数据收集是无线传感网络研究的关键问题,是诸多无线传感网络应用的基础.降低数据聚集的延迟是数据聚集研究中的重点问题.现有的面向延迟的数据聚集算法,多是通过在树型网络结构上设计无冲突的节点调度算法,来降低数据聚集的延迟,没有考虑到无线网络数据易丢失的特性,不能达到期望的延迟效果.本文针对上述问题,提出一种采用任意播(anycast)方式的机会数据收集算法(OA算法).该算法利用机会传输(opportunistic transmission)的思想,用任意播方式传输数据,通过减少数据聚集中重传数据包的数目,来降低数据聚集的延迟.实验表明,与SPT(Shortest path tree)上的数据聚集延迟相比,该方法的发包数目减少了15%,延迟降低了10%.Data Aggregation is a key problem in wireless sensor networks.It has a wild of applications in environment monitoring and scientific observation.In this paper,we focus on shortening the latency of data aggregation.Most of previous works usually schedule nodes on a tree rooted at the sink node by a collision free scheduling.As wireless channel is unstable and message loss always happens,the scheduling on the fixed tree structure cannot achieve a good performance on latency.To overcome this problem,we design an opportunistic data aggregation algorithm using anycast,named OA algorithm.As less packets are transmitted,lower latency is achieved.We validate our algorithm by simulation.Compared with scheduling algorithm on shortest path tree（SPT）,the results show that about 15% less messages are transmitted by OA and the latency by OA is 10% lower than that by SPT.

关 键 词：无线传感网络 数据聚集 机会数据聚集 任意播 

分 类 号：TP309[自动化与计算机技术—计算机系统结构]