基于二分K-means的无线传感器网络分簇方法  

Research on clustering method of wireless sensor network based on bisecting K-means

作　　者：张本宏[1] 江贺训 ZHANG Benhong;JIANG Hexun(School of Computer and Information,Hefei University of Technology,Hefei 230601,China)

机构地区：[1]合肥工业大学计算机与信息学院

出　　处：《合肥工业大学学报（自然科学版）》2020年第1期39-44,123,共7页Journal of Hefei University of Technology：Natural Science

基　　金：国家自然科学基金资助项目(61370088)

摘　　要：好的分簇方法可以通过有效提高网络能量利用率均衡网络负载延长网络生命周期,文章提出一种基于二分K-means的均匀分簇算法(uniform clustering optimization algorithm,UCOA)。该算法首先基于对网络能耗的理论分析确定网络最优簇头数目,然后基于最优簇头数目利用二分K-means算法对整个网络均匀分簇,加入节点剩余能量和距离因子改进簇头选举阈值公式,并且在簇头与基站通信时采用单跳和多跳相结合的数据传输方式。仿真实验表明UCOA分簇算法能有效提高节点耗能均衡性,延长网络生存时间。Good clustering method can effectively improve the network energy utilization,balance network load and prolong the life cycle of network.A uniform clustering optimization algorithm(UCOA)based on bisecting K-means is proposed.The algorithm first determines the number of network optimal cluster heads based on the theoretical analysis of network energy consumption,and then uses the optimal cluster head number to divide the whole network evenly with the bisecting K-means algorithm,and improves the threshold formula of the cluster head election by adding the node residual energy and distance factor,and uses the combination of single-hop and multi-hop data transmission when the cluster head communicates with the base station.The simulation results show that UCOA clustering algorithm can effectively improve the energy consumption balance and prolong the network lifetime.

关 键 词：无线传感网络(WSN) 最优簇头数 二分K-means 均匀分簇 

分 类 号：TP212.9[自动化与计算机技术—检测技术与自动化装置]