基于K-means的无线传感器网络分簇算法研究

摘要

提出一种基于K-均值聚类的无线传感器网络分簇算法.从K-均值聚类算法中要解决的合理聚类数的确定、初始聚类中心的选择以及聚类性能对目标函数的依赖这三个问题入手,运用K-均值聚类算法来实现无线传感器网络分簇.仿真与性能分析结果表明,基于K-均值聚类的无线传感器网络分簇算法既能节省节点能量、延长网络生命,又能改善网络中的能耗均衡,并保证簇首分布的均匀性.%A clustering algorithm based on K-means clustering for WSN is proposed. Since the questions of the determining number of the best clusters, the choice of initial cluster centers and objective function are settled, then the sub-clusters of WSN can be achieved with the K-means clustering algorithm. Finally, evaluating the sub-cluster algorithm performance and simulating it which shows that the WSN clustering algorithm based on the K-means clustering can not only save energy and prolong the life of the network, but also improve the energy consumption of a balanced network and ensure uniform distribution of the clustering heads.