基于Q学习的多基站分簇拓扑控制算法  

Clustering Topology Control for Multiple Base Stations in WSNs with Q-Learning

作　　者：阎新芳[1] 冯岩[1] 王晓晓[1] 

机构地区：[1]郑州大学信息工程学院,郑州450001

出　　处：《传感技术学报》2016年第4期578-582,共5页Chinese Journal of Sensors and Actuators

基　　金：河南省科技厅基础与前沿研究计划项目(152300410023)

摘　　要：为了解决无线传感器网络中单基站附近出现的"能量空洞"和网络时延过高等问题,引入多基站分簇拓扑控制算法。算法根据不同的场景来选择基站数目,结合图论和定向扩散中梯度的思想对网络进行分簇并运用Q学习算法对簇头节点进行周期性的学习训练,比较到达不同基站的不同路径上的Q值进行最优路径的选择。通过仿真分析表明,该算法相对于单基站分簇算法可以有效延长网络的生命周期。This paper presents a multiple base stations clustering topology control algorithm to solve the issue of＂en-ergy hole＂and high network delay near a single base station in wireless sensor networks（WSNs）. The number ofbase stations is determined according to different scenarios. Our algorithm uses a method of graph theory and the gra-dient of directional diffusion to clustering. To achieve the clustering topology control,our method exploits Q-learningmethod,incrementally learning at each node＇s sufficient network knowledge to choose the best path to base stations.Evaluation of the resulting our algorithm demonstrates its ability to significantly increase the lifetime of network incomparison to the single base station clustering algorithm.

关 键 词：无线传感器网络 分簇拓扑控制 多基站 Q学习 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]