基于模拟退火算法的传感器网络聚类方案  

Clustering scheme based on simulated annealing algorithm in wireless sensor network

作　　者：胡荣[1] 杨春[2] 何军[3] 李奇[2] 

机构地区：[1]四川师范大学计算机科学学院,成都610068 [2]四川师范大学四川省计算机软件重点实验室,成都610068 [3]四川大学计算机学院,成都610064

出　　处：《计算机应用》2010年第2期299-302,共4页journal of Computer Applications

基　　金：国家863计划项目(2008AA01Z208);四川省科技厅基金资助项目(2007H12-010);四川省计算机软件重点实验室资助项目(08JSU002);四川省青年基金资助项目(2009-28-419)

摘　　要：针对传感器网络聚类间能耗负载不均衡而引发的"能量热点"问题,提出一种在节点随机分布情况下构建能量负载均衡的聚类方法。网络中高于平均能量的节点率先成为候选首领,候选首领根据剩余能量、节点连通度以及当选首领的总时间来竞争聚首,普通节点首先依据信号强弱选择加入聚类,然后借鉴模拟退火算法动态调整所拥有的成员节点,直到所有聚类的能耗状态趋近均衡。仿真表明,与基于均匀分布假设的聚类方案相比,新方案具有能耗负载更均衡的聚类和更长的生命周期。Targeting at the problem of Energy Hotspot caused by inter-cluster disadvantages of disequilibrium energy consumption load in Wireless Sensor Network （WSN）, a clustering scheme of balanced energy load was proposed in the situation of randomly distributed nodes. The nodes whose energy was higher than the average energy of the network became the first candidate heads. Then the candidate heads competed to be clustering head according to their residual energy, connectivity, as well as the total time of being ever clustering head. The general node selected a cluster to join based on signal strength received from the clusters, and in accordance with the idea of simulated annealing algorithm, dynamically adjusted the members in each cluster, until each cluster approximated a balanced state of energy consumption. Simulation results show that the new scheme can obtain more balanced energy load clusters and longer life time than those based on uniform distribution

关 键 词：无线传感器网络 模拟退火算法 能耗均衡 聚类 

分 类 号：TP393[自动化与计算机技术—计算机应用技术] TN929.5[自动化与计算机技术—计算机科学与技术]