基于Agent的无线传感器网络分簇拓扑控制方法研究.pdf

南京邮电大学
硕士学位论文
基于Agent的无线传感器网络分簇拓扑控制方法研究
姓名:桂鹤飞
申请学位级别:硕士
专业:计算机应用技术
指导教师:章韵
2011-01
南京邮电大学硕士研究生学位论文摘要
摘要
无线传感器网络是由大量低成本、具有传感、数据据处理的无线传感器节点通过自组织
方式组成的网络。传感器节点一般采用电池供电,且工作在较为复杂、危险的区域,传感
器节点得不到到能量的补充。高效节能的拓扑控控制制算法可以有效的降低节点能耗和和延长网络
生存时间。
论文在多 Agent 系统的基础上提提出了两种适于移动动汇聚节点无线传感器网络的自适应
分簇拓扑控控制制算法,分别是基于多 Agent 系统的加加权自适应拓扑控控制制算法( STCAW,
Self-adaptive Topology Control Algorithm based on Weight)和和基于合适簇规模的自适应拓扑
控控制制算法(SCACS, Self-adaptive topology Control algorithm based on Appropriate Clustering
Scale) 。STCAW 算法通过设定权值,将节点 Agent 的剩剩余能量,邻居数目和和与汇聚 Agent
之间的距离作为簇首 Agent 的选择依据据,选择最优的节点 Agent 成为簇首 Agent 。SCACS
算法首先通过 STCAW 算法选出最优的节点 Agent 作簇首 Agent ,再通过功功率控控制制调整簇
首 Agent 的覆盖半径使簇的规模落在合理的范围之内,通过分簇得到到无线传感器网络的最
优簇结构。在网络的分簇结束以后,SCACS 算法保证所有节点 Agent 都在簇首 Agent 的覆
盖范围内。
论文采用++ 对提提出的改进算法进行了仿真。仿真结果表明,改进的算法可以
有效的提提高网络效率和和延长网络的生命命周周期。


关键词关键词::::无线传感器网络,拓扑控控制制,分簇,移动动汇聚节点
I
南京邮电大学硕士研究生学位论文 ABSTRACT
ABSTRACT
The wireless works , composed of lots of low-cost nodes with abilities of sensing ,
processing data and munication, is works. The energy of the nodes
is supplied by the batteries. Because the nodes work in the dangerous regions, the energy of the
nodes cannot be supplied easily. So, reducing the consumption of the energy and prolonging the
lifetime of works is very important.
Based on the mutli-agent system, this dissertation proposes STCAW (Self-adaptive Topology
Control Algorithm based on Weight)and SCACS( Self-adaptive topology Control algorithm based
on Appropriate Clustering Scale). By setting weights, inluding the residual energy of the agent,
the average distances between agent and the sink agent, the count of the agent’s neighbors,
STCAW chooses the best agent as the cluster header. Firstly, by the STCAW algorithm, SCACS
chooses the best agent as the cluster head. Then, by adjusting the transmit power of the cluster
head, the work clustering will be obtained. After forming the work
clustering, SCACS ensures th