基于频繁项挖掘的空间关联性子簇形成算法  

Frequent Itemset Mining-Based Spatial Subclustering Algorithm

作　　者：王茜[1] 高志鹏[1] 邱雪松[1] 王兴斌 

机构地区：[1]北京邮电大学网络与交换技术国家重点实验室,北京100086 [2]总参信息化部驻成都地区军事代表室,成都610041

出　　处：《北京邮电大学学报》2015年第B06期20-23,共4页Journal of Beijing University of Posts and Telecommunications

基　　金：国家自然科学基金项目(61272515;61372108;61121061);北京高等学校青年英才计划项目(YETP0474);教育部博士点基金项目(20110005110011)

摘　　要：部署在无线传感器网络监测区域的传感器节点周期性地进行感知数据的采集和传输,传感器节点采集数据之间存在的空间关联性会增加采集数据的冗余度和网络能耗.为了延长无线传感器网络的生命周期,提出了一种基于频繁项挖掘的空间关联性子簇形成算法.仿真实验结果表明,该算法与已有算法相比,降低了网络能耗,延长了网络的生命周期,保证了采集数据的质量.The cluster-based periodic data collection for wireless sensor networks, the sensor nodes de- ployed in monitoring region periodic transmit data to sink nodes were considered. The spatial correlation in collected data increases the redundancy of data and network energy consumption. An algorithm was proposed to prolong the lifetime of wireless sensor network and ensure the fidelity of collected data. A fre- quent itemset mining-based spatial subclustering algorithm was proposed. Analysis and experiment show that the improved algorithm can achieve more energy savings, extend the wireless sensor networks lifetime and guarantee the fidelity of collected data.

关 键 词：无线传感器网络 数据采集 频繁项挖掘 空间关联性 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]