摘要
Wi-Fi指纹匹配定位算法具有简单、快捷、方便、经济、易普及等诸多优点,但对位置指纹的匹配精度较低。对此,提出一种贝叶斯与加权K近邻算法相结合的贝叶斯概率优化算法,应用于Wi-Fi指纹匹配定位,在提高传统加权K近邻算法精度的同时,减少了贝叶斯概率匹配算法的平均运行时间。实验结果显示,该算法可以将1 m内的定位精度从原先的57%提升至73%,平均定位精度提高约21.49%,定位稳定性也有所加强。
Fingerprint matching positioning method of Wi-Fi offers simple,convenient,ubiquitous,and economic solutions for indoor positioning services,but the setback of this method is its low precision.This paper proposes a Bayesian probabilistic optimization algorithm combining Bayesian and weighted K-nearest neighborhood algorithm,which is applied to Wi-Fi fingerprint matching positioning.The proposed combination of Bayesian with weighted K-neighborhood algorithm improves the accuracy and reduces the average running time.The simulation shows that proposed algorithm improves the positioning accuracy from 57%to 73%,with an average of about 21.49%,and the positioning stability is also enhanced.
作者
杨如民
陈敏
余成波
Yang Rumin;Chen Min;Yu Chengbo(Institute of Remote Test and Control,Chongqing University of Technology,Chongqing 400054,China)
出处
《计算机应用与软件》
北大核心
2021年第2期97-102,144,共7页
Computer Applications and Software
基金
国家高端外国专家项目(GDW20185200480)
重庆市科技人才培养计划项目(CSJC2013KJRC-TDJS40012)
重庆市高校优秀成果转化资助项目(KJZH14213)
关键词
指纹定位
加权K近邻算法
指纹库
贝叶斯概率算法
Fingerprint positioning
Weighted K-Nearest neighborhood
Fingerprint database
Bayesian probability algorithm
