文献类型：期刊文献

中文题名：基于参考位置指纹离散程度的WKNN定位方法

英文题名：WKNN positioning method based on fingerprint dispersion degree of reference location

作者：赵芳[1]

第一作者：赵芳

机构：[1]新乡学院计算机与信息工程学院,河南新乡453003

第一机构：新乡学院计算机与信息工程学院

年份：2021

卷号：43

期号：2

起止页码：203-207

中文期刊名：沈阳工业大学学报

外文期刊名：Journal of Shenyang University of Technology

收录：CSTPCD;;Scopus;北大核心:【北大核心2020】；

基金：河南省科技厅基金项目(B20143369).

语种：中文

中文关键词：无线网络节点定位;指纹定位;位置服务;无线定位技术;离散程度;K-means算法;数据聚类;WKNN定位

外文关键词：wireless network node positioning;fingerprint positioning;location service;wireless positioning technology;dispersion degree;K-means algorithm;data clustering;WKNN positioning

摘要：针对目前在室内复杂环境内Wi-Fi无线接入点位置指纹定位精度低且不同位置时定位精度波动较大等问题,提出了基于参考位置指纹离散程度的WKNN定位方法.构建初始位置指纹数据库,基于K-means算法完成数据聚类.采用DD-WKNN指纹定位算法选取与RSSI值欧氏距离最相近的指纹数据,根据离散程度估算参考权重的权值,完成归一化加权求和,估测得到待测位置的坐标.结果表明,该算法有效提高了定位精度,同时误差波动较小,实用性较强.
Aiming at the problems of low fingerprint positioning accuracy with Wi-Fi wireless access points and large fluctuation of positioning accuracy at different locations within indoor complex environment,a WKNN positioning method based on the fingerprint dispersion degree of reference location was proposed.An initial location fingerprint database was constructed,and the data clustering based on the K-means algorithm was completed.The DD-WKNN fingerprint positioning algorithm was used to select the fingerprint data with the closest RSSI value Euclidean distance.The weight number of reference weight was estimated according to the dispersion degree,the normalized weighting summation was completed,and the location coordinates to be measured were estimated.The results show that the as-proposed algorithm effectively improves the positioning accuracy,and has small error fluctuation and better applicability.