基于XGBoost的客户所在店铺WiFi定位技术研究  

A Study on Store Positioning Technology through Customers’WiFi Location Based on XGBoost

作　　者：卢一帆 柳伟[1] 叶福田 Lu Yifan;Liu Wei;Ye Futian(National Engineering Research Center of Die and Mold CAD, Shanghai 200030, China;Hengli Mold Technology Development Co.Ltd., Dongguan523460, China)

机构地区：[1]上海交通大学模具CAD国家工程研究中心,上海200030 [2]东莞市横沥模具科技产业发展有限公司,广东东莞523460

出　　处：《计算机测量与控制》2019年第7期141-145,共5页Computer Measurement &Control

基　　金：国家科技重大专项(2017ZX04016001)

摘　　要：为了减小复杂室内环境等因素对WiFi定位的影响,降低定位成本,提高定位精度并缩小定位区域,通过对室内定位技术和相关机器学习算法进行深入分析和探讨,提出了一种基于XGBoost的WiFi室内定位算法;根据WiFi信号强度分布非均匀的特点,通过提取移动端WiFi强度特征,并利用XGBoost分类器对信号来源进行定位;实验结果表明,该定位算法在WiFi强度特征可检测时达到了87.72%的定位精度,达到预期的定位效果,同时定位时间较短,稳定性较好,可以基本满足实时定位的要求。In order to lessen the influence of indoor complex environment factors on WiFi positioning, reduce the positioning cost, improve the positioning accuracy and lock the positioning area, a WiFi indoor location algorithm based on XGBoost is proposed through in-depth analysis and discussion of the indoor positioning system and related machine learning algorithms. According to the non-uniform characteristics of WiFi signal strength distribution, this algorithm extracts WiFi intensity features and uses XGBoost to locate the signal source. Experimental results show that the positioning algorithm achieves 87.72% positioning accuracy when detecting the WiFi intensity feature, which achieves the desired positioning effect, with short positioning time, good robustness, and can meet the requirements of real-time positioning.

关 键 词：室内环境 手机定位 分类器 位置指纹 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置]