工业WSNs中基于Q-学习的图路由算法  

A Q-Learning-Based Graph Routing in Industrial Wireless Sensor Networks

作　　者：罗坤[1] 赵新颖[2] LUO Kun;ZHAO Xinying(Zhengzhou Railway Vocational and Technical College,International Education College,Zhengzhou He'nan 451460,China;Zhengzhou Railway Vocational and Technical College,School of electronic engineering,Zhengzhou He'nan 451460,China)

机构地区：[1]郑州铁路职业技术学院国际教育学院,河南郑州451460 [2]郑州铁路职业技术学院电子工程学院,河南郑州451460

出　　处：《传感技术学报》2020年第10期1496-1501,共6页Chinese Journal of Sensors and Actuators

摘　　要：针对工业无线传感网络(Industrial Wireless Sensor Networks,IWSN),提出基于Q-学习的图路由(Q-Learning-based Graph Routing,QLGR)。利用图表述网络拓扑,QLGR算法通过网络内节点的信息构建上行链路图。先依据节点距网关的跳数,能量供应类型以及离邻居节点接收信号强度构建节点的成本函数,再依据成本函数选择节点加入上行链路图。同时,利用Q-学习算法调整成本函数的权重系数,进而减少数据传输时延,延长网络寿命。仿真结果表明,提出的QLGR算法减缓了节点能量消耗速度,提高了传输数据的可靠性。For Industrial Wireless Sensor Networks(IWSN),Q-learning-based Graph Routing(QLGR)is proposed in this paper.The network topology is represented by a graph,and the QLGR algorithm uses to the information of nodes to construct the uplink graph.The cost function of the node is first constructed according to the number of hops of the access point of the node,the type of supplied energy and the received signal strength of the neighboring node,and then the node is selected to join the uplink diagram according to the cost function.At the same time,Q-learning algorithm is used to adjust the weight coefficient of the cost function,so as to reduce the data transmission delay and extend the network life.Simulation results show that the proposed QLGR algorithm can slow down the node energy consumption speed and improve the reliability of data transmission.

关 键 词：工业无线传感网络 图路由 Q-学习 上行链路图 成本函数 

分 类 号：TPT393[自动化与计算机技术]