基于机器学习的无线传感网络室内定位研究  

Research on Indoor Location of Wireless Sensor Network Based on Machine Learning

作　　者：郑春 张继山[1] ZHENG Chun;ZHANG Jishan(Anhui Sanlian University,Hefei 230601,China)

机构地区：[1]安徽三联学院,安徽合肥230601

出　　处：《太原师范学院学报（自然科学版）》2020年第1期49-53,64,共6页Journal of Taiyuan Normal University:Natural Science Edition

基　　金：2018年安徽省教育厅自然科学重点项目(基于机器学习的无线传感网络室内定位方法及应用研究)(KJ2018A0602);2018年度安徽三联学院计算机网络技术重点实验室专项研究项目成果《基于网络行为分析的网络管理系统设计与研究》(PTZD20180003);2017年度安徽三联学院服务机器人协同创新中心专项研究项目成果《复杂环境下城市智能交通系统空车位导引关机技术研究》(ZJQR17002).

摘　　要：无线传感网络在能耗量化传导过程中具有随机分布性,导致网络节点的室内定位精度不高,为了提高网络的室内定位准确性,提出基于机器学习的无线传感网络室内定位方法.构建无线传感网络室内定位的节点优化部署模型,采用能量负载均衡控制方法进行无线传感网络的路由探测协议设计,建立无线传感网络节点传输的链路均衡配置模型,采用机器学习算法进行无线传感网络室内定位过程中的自适应寻优,提取无线传感网络节点输出信号的能谱特征量,根据能谱的聚类属性进行无线传感网络室内定位优化.仿真结果表明,采用该方法进行无线传感网络室内定位的精度较高,能量开销较小,网络节点的组网部署能力得到提升.Wireless sensor networks(WSNs)have random distribution in the process of quantitative transmission of energy consumption,which leads to low indoor positioning accuracy of network nodes.In order to improve the indoor positioning accuracy of the network,a machine learning-based indoor positioning method for wireless sensor networks is proposed.The optimal deployment model of wireless sensor network indoor positioning node is constructed.The routing detection protocol of wireless sensor network is designed by using the energy load balancing control method,and the link balance configuration model of wireless sensor network node transmission is established.The machine learning algorithm is used for adaptive optimization in wireless sensor network indoor location.The energy spectrum characteristic of the output signal of wireless sensor network node is extracted and the indoor location optimization of wireless sensor network is carried out according to the clustering attribute of energy spectrum.The simulation results show that the indoor positioning accuracy of the wireless sensor network is high,the energy cost is small,and the network deployment ability of the network nodes is improved.

关 键 词：机器学习 无线传感网络 室内定位 路由探测 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]