无线传感网中分簇分层k-medoids协议研究

研究了大规模高密度节点部署的无线传感器网络(Wireless Sensor Networks),提出了一种具有可扩展性的分层分簇k-medoids协议.在簇的建立阶段采用改进的k-medoids聚类算法分簇,并将多级分层技术和k-medoids分簇算法融合形成多个层次的分层分簇网,在层0完成全部分簇之后,启动第1层的分簇,并以此类推直到用户指定的层数.仿真实验表明,多层k-medoids分簇算法缩减了网络节点与sink节点之间的通信量,降低了能耗,最终延长了网络生存期.

A study of the distributed layered k-medoids clustering protocol in wireless sensor networks

In order to solve the problem of the large- scale and high- density wireless sensor networks,an adaptive distributed hierarchical k- medoids clustering protocol is proposed. In the establishment stage,a multilevel hierarchical clustering network integrating the improved k- medoids clustering algorithm with the hierarchical clustering technology is constructed; after the layer 0 completes all the clustering,it starts the clustering of the first layer,and then the next until the user- specified number. The simulation experiments show that this multilayer k- medoids clustering algorithm can reduce not only the amount of communication between the network nodes and sink nodes but also energy consumption,and thus prolongs the network survival period.