地理空间信息2024，Vol.22Issue(8)：102-106,5.DOI:10.3969/j.issn.1672-4623.2024.08.023
基于增强指纹特征的WiFi室内定位方法
WiFi Indoor Location Method Based on Enhanced Fingerprint Feature
摘要
Abstract
Fingerprint-based WiFi location is one of the commonly used indoor location methods,but the WiFi location fingerprint database has problems,such as wireless access point(AP)distribution and sparse samples,which seriously affect the location effect in indoor local areas.For this reason,we proposed a WiFi indoor location method based on enhanced fingerprint feature.Firstly,at the stage of building a fingerprint data-base,we used AP preferential selection strategy with joint reception percentage and variance to screen out more stable and better quality AP points.Then,we expanded the fingerprint database in the corresponding location,so as to reduce the workload of establishing the fingerprint data-base in the early stage and improve the quality of fingerprint database.Finally,at the stage of matching location,we fused the Pearson correlation coefficients to construct a weighted k Nearest Neighbor(kNN)location algorithm.Test results of actual location scenarios show that the location accuracy is improved by 22.5%using AP-preferred fingerprint database and the weighted kNN location algorithm incorporating Pearson correla-tion coefficients.The method can effectively improve the location accuracy while reducing the cost of capturing fingerprint databases.关键词
指纹特征增强/稀疏样本/相关系数/WiFi室内定位Key words
fingerprint enhancement/sparse sample/correlation coefficient/WiFi indoor location分类
天文与地球科学引用本文复制引用
宁夏薇,夏林元,祝宏宇,刘俊求..基于增强指纹特征的WiFi室内定位方法[J].地理空间信息,2024,22(8):102-106,5.基金项目
广东省重大专项资助项目(2015B010104003) （2015B010104003）
广州市产学研协同创新重大专项资助项目(201604046007). （201604046007）
