无线传感器网络中高能力簇首节点部署问题  

High-Power Cluster-Head Deployment in Wireless Sensor Networks

作　　者：刘琳[1] 黄艳 于海斌[3] 

机构地区：[1]国家卫星海洋应用中心,北京100081 [2]苏州市职业大电子信息工程系,江苏苏州215021 [3]中国科学院沈阳自动化研究所,沈阳110016

出　　处：《传感技术学报》2010年第7期1023-1029,共7页Chinese Journal of Sensors and Actuators

基　　金：国家自然科学基金重点项目资助(60434030);国家杰出青年基金项目资助(60725312)

摘　　要：分簇对用于环境监测的无线传感器网络具有较好的适应性和节能性,由高能力节点担任簇首可以更好的实现节能并改善网络性能,从而延长网络生命期。当网络规模较大时,高能力簇首节点的部署问题是一类NP-hard问题。本文首先对此问题进行了形式化描述,进而分析了跳数与网络性能之间的关系,总结出跳数是影响网络能耗和报文传输实时性的主要影响因素,簇首的部署应使簇成员和簇首之间的跳数尽可能小。因此提出了一种基于K-平均的簇首部署策略(KMCD),通过有策略的部署少量簇首节点来实现网络性能的最优化。仿真分析表明,与现有算法相比,KMCD算法具有较好的节能性和实时性。Cluster-based method has better adaptability and energy-efficiency to multi-hop Wireless Sensor Networks （WSNs） used for environmental monitoring. If the Cluster Head （CH） is served by more powerful node, the WSN performance will be improved greatly, thereby the network lifetime will be prolonged. When the WSN scale is big enough the problem of how to deploy the high-power CH is NP-hard problem. Firstly, the deployment problem is formulated as an integer linear programming problem in this paper. Then, the relationship is analyzed between hop-count and network performance. The analysis suggests that hop-count is the main factor which greatly affects the performance of WSN energy consumption and real time data transmission and the CH deployment should enable the hop-count between CH and its member as small as possible. Finally, the paper puts forward a CH deployment strategy based on K-mean algorithm （KMCD） which can optimize the network performance by strategically deploying a few CHs. Simulation results show that KMCD can efficiently improve energy efficiency and real-time packet transmission compared with the existing methods.

关 键 词：无线传感器网络 分簇 K-平均 选址问题 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]