基于阈值筛选模糊分簇的WSN数据稳定汇聚算法  

WSN data stability convergence algorithm based on threshold selection of fuzzy clustering

作　　者：谷川[1] 张志彦[1] 开金宇[2] 

机构地区：[1]安阳师范学院软件学院,河南安阳455000 [2]上海大学计算机与信息工程学院,上海200444

出　　处：《计算机工程与设计》2016年第9期2315-2320,共6页Computer Engineering and Design

基　　金：国家自然科学基金项目(60875081);河南省教育厅自然科学研究计划基金项目(2009B520001)

摘　　要：为改善当前WSN数据稳定汇聚过程中难以动态筛选最佳簇头节点及数据传输过程中节点内部竞争过度导致数据传输抖动的问题,提出一种基于阈值筛选模糊分簇的无线传感网数据稳定汇聚算法。在分簇过程中基于模糊思想形成分簇阈值,根据阈值水平高低进行网络结构的初始化,利用剩余能量水平进行簇头节点的周期轮换;在簇头节点确立后,综合考虑剩余能量及与Sink节点间物理距离,定义数据汇聚准则,簇成员通过该准则将数据传输至控制中心节点,完成整轮数据稳定汇聚过程。仿真结果表明,与当前WSN数据汇聚算法相比,该算法可改善网络负载水平,增强网络数据分组投递性能,减少网络控制开销。It is difficult to dynamically select the optimal cluster head node during the current WSN data convergence,and data transmission jitters induced by excessive internal competition of nodes in data transmission.To solve these problems,the WSN data stability convergence algorithm based on threshold selection of fuzzy clustering was proposed.The clustering threshold was formed based on the fuzzy theory,and the network structure was initialized according to the threshold level of the network structure,and the periodic rotation of the cluster head was carried out using the residual energy level.The data convergence criterion was defined by considering the residual energy and the physical distance between the Sink nodes after establishing the cluster head node,and the cluster member transfered the data to the control center node complied with this criterion for completing the stable convergence of whole data.Simulation results show that compared with the traditional clustering algorithm,the proposed algorithm can effectively improve the network load level,enhance network data packet delivery performance,and reduce network control overhead.

关 键 词：无线传感器网络 数据汇聚 模糊分簇 阈值筛选 周期轮换 剩余能量 

分 类 号：TP393.04[自动化与计算机技术—计算机应用技术]