摘要
针对Wi-Fi信号强度的相似性对室内定位的影响,本文提出一种基于Wi-Fi指纹和随机森林的室内定位算法。该算法采用Wi-Fi作为信号源,以接收信号强度指示和基本服务集标识符来构建Wi-Fi指纹库,从而建立随机森林模型用于室内位置感知。仿真实验表明,该算法的定位误差约为2.26 m,与同类算法相比,在执行时间和定位精度上具有较好的优越性,算法精度提高约3.2%。
Aiming at the influence of the similarity of Wi-Fi signal strength on indoor positioning,this paper proposes an indoor positioning algorithm based on Wi-Fi fingerprint and Random Forest.The algorithm uses Wi-Fi as a signal source to construct the Wi-Fi fingerprint library by receiving signal strength indication and basic service set identifiers,thereby establishing a Random Forest model for indoor location sensing.The simulation results show that the positioning error of the algorithm is about 2.26 m.Compared with similar algorithms,it has better performance in execution time and positioning accuracy,and the accuracy of the algorithm is improved by about 3.2%.
作者
韩学法
吴飞
时瑶佳
胡锐
聂大惟
HAN Xuefa;WU Fei;SHI Yaojia;HU Rui;NIE Dawei(School of Electronic and Electrical Engineering,Shanghai University of Engineering Science,Shanghai 201620,China)
出处
《智能计算机与应用》
2021年第7期102-105,112,共5页
Intelligent Computer and Applications
基金
上海市高校研究生创新项目(19KY0216)
上海市科技学术委员会重点项目(18511101600)
国家科学基金青年基金项目(61902237)
上海市科委青年科技英才“扬帆计划”项目(19YF1418200)。
关键词
室内定位
Wi-Fi指纹
指纹数据库
随机森林
位置感知
indoor positioning
Wi-Fi fingerprint
fingerprint database
Random Forest
location perception