基于K-means++的无线传感网能量高效分簇协议  

An energy-efficient clustering protocol based on K-means++ for the wireless sensor network

作　　者：余扬[1] 李艳彩 陈君华[1] YU Yang;LI Yan-cai;CHEN Jun-hua(Key Laboratory of IOT Application Technology of Universities in Yunnan Province,Yunnan Minzu University,Kunming 650500,China)

机构地区：[1]云南民族大学云南省高校物联网应用技术重点实验室,云南昆明650500

出　　处：《云南民族大学学报（自然科学版）》2019年第2期165-171,共7页Journal of Yunnan Minzu University:Natural Sciences Edition

基　　金：云南民族大学物联网工程虚拟仿真实验教学中心(2017XNFZ01);云南民族大学研究生创新基金(2018YJCXS229)

摘　　要：针对传统无线传感网的分层路由算法中存在着分簇不均匀、簇首数量不固定、簇首位置不合理、节点的可扩展性不足以及数据传输方式比较单一的问题,提出一种无线传感网能量高效分簇协议.该协议在簇的建立阶段基于K-means++聚类算法进行分簇并采用S_Dbw聚类评价指标挑选其最优分簇,在簇的建立阶段,从每个簇中选取簇内剩余能量最高的节点作为簇首;在数据传输阶段,基于节点间的通信代价使用Dijkstra算法来寻找每个簇首到汇聚节点的最优路径.仿真结果表明:该协议可降低节点与汇聚节点之间数据传输的能耗,延长传感网的生命周期,并且在整个网络能量处于较低水平时也可以较好的覆盖整个监测区域.The hierarchical routing algorithm for the traditional wireless sensor network has the problems of uneven clustering,an unfixed number of cluster heads,improper positions of cluster heads,insufficient node scalability and single-hop data transmission. Aiming at solving the above problems,a clustering protocol for energy-efficient clustering of wireless sensor networks is proposed. In the cluster establishment stage,a K-means++ clustering algorithm is used for clustering. The clustering evaluation S_Dbw index is used to select the optimal clustering result. In the cluster establishment stage,the node with the highest remaining energy in each cluster is selected as the cluster head. In the data transmission phase,it uses the Dijkstra algorithm to find the optimal path from each cluster head to the sink node based on the communication cost. Simulation results show that this algorithm saves the energy consumption of data transmission between the network node and the base station,prolongs the life cycle of the network,and can cover the entire monitoring area when the energy of the whole network is low.

关 键 词：K-means++算法 S_Dbw 自适应分簇 可扩展性 DIJKSTRA算法 

分 类 号：TP212.9[自动化与计算机技术—检测技术与自动化装置]