Ad Hoc网络CF分簇算法  

Connectivity First Clustering Algorithm for Ad Hoc Networks

作　　者：夏利[1] 马学文[1] 刘阳迁[1] 刘式舟[1] 

机构地区：[1]东北大学信息科学与工程学院,辽宁沈阳110004

出　　处：《沈阳建筑大学学报（自然科学版）》2008年第3期516-519,共4页Journal of Shenyang Jianzhu University：Natural Science

基　　金：国家自然科学基金项目(60773219);辽宁省自然科学基金资助项目(20042012)

摘　　要：目的提出一种新的Ad Hoc网络连通性优先(CF)分簇算法,优先保证区域的连通性.方法在HD算法的基础上进行改进,在区域划分的时候加入连通性的限制条件,CF算法将节点划分为成员节点,可选节点和自由节点3种状态,采取逐步扩张的思想不断将可选节点和自由节点转化为成员节点,在保证区域连通性的前提下完成簇的划分.结果在节点密度较高的情况下,CF与HD均能保证整个区域是连通的,随着节点密度的下降,CF算法仍然能保证区域的连通性,HD则出现了不能与簇外区域通讯的簇.两种算法在相同节点密度下,得到的簇数量接近.结论在节点密度较高或较低的情况下,CF算法均能较好的保证区域连通性.Due to transmitting power and channel restrictions, nodes need to cluster in wireless Ad Hoc networks. Clustering algorithm plays a key role in the performance of ad hoc networks. The goal of existing Highest Degree algorithm is to minimize the number of clusters. It can＇t guarantee connectivity. This paper presents a new algorithm called Connectivity First Clustering Algorithm which is based on HD Algorithm, and considers connectivity priority. In CF algorithm, nodes are divided into three types, such as member nodes, option nodes and free nodes. Gradually, free nodes change to member nodes or option nodes through expansion strategy. It＇s proved by evaluation that it can maintain domain connectivity well under the circumstances of low density.

关 键 词：Ad HOC 分簇算法 连通性 节点密度 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]