一种层次Levenshtein距离的无指纹校准的室内定位方法

发布时间：2019-02-27 06:32

【摘要】：随着移动计算领域的兴起,基于位置的服务越来越受青睐。目前各种室内定位的方法层出不穷,由于室内广泛部署了无线基础设施,基于WiFi指纹信息的室内定位技术是其主流方法。设备异构和室内环境变化是影响定位精度的主要因素。本文针对以上两个问题,提出一种层次Levenshtein距离(HLD)的WiFi指纹距离计算算法,实现异构设备的指纹无校准比对。将不同移动设备采集的RSSI信息转化为AP序列,根据AP对应的RSSI值的差异性计算其层次能级,结合Levenshtein距离计算WiFi指纹之间的距离。对于需定位的WiFi指纹RSSI信息,利用HLD算法获取K个近邻,采用WKNN算法进行预测定位。实验中,为了验证算法的鲁棒性和有效性,在3种不同类型的室内环境中采用5种不同的移动设备来采集WiFi的RSSI信息,其定位的平均精度达1.5 m。
[Abstract]:With the rise of mobile computing, location-based services are becoming more and more popular. At present, various indoor location methods emerge one after another. Because of the widespread deployment of wireless infrastructure, indoor location technology based on WiFi fingerprint information is the mainstream method. Heterogeneous equipment and indoor environment change are the main factors affecting positioning accuracy. In order to solve the above two problems, this paper proposes an algorithm for calculating the WiFi fingerprint distance of hierarchical Levenshtein distance (HLD), which can realize the fingerprint calibration-free comparison of heterogeneous devices. The RSSI information collected by different mobile devices is transformed into AP sequence. The hierarchical energy levels are calculated according to the difference of RSSI value corresponding to AP, and the distance between WiFi fingerprints is calculated by combining Levenshtein distance. For the RSSI information of WiFi fingerprints which need to be located, K nearest neighbors are obtained by using HLD algorithm, and WKNN algorithm is used to predict and locate. In order to verify the robustness and effectiveness of the algorithm, 5 different mobile devices are used to collect the RSSI information of WiFi in three different indoor environments. The average positioning accuracy is 1.5m.
【作者单位】：皖西学院电子与信息工程学院;安徽大学智能计算与知识工程研究所;清华大学软件学院可信网络与系统研究所;
【基金】：国家自然科学基金项目(61572366,61303209,61522110,61402006,61673020) 2016年安徽省高校优秀中青年骨干人才国内外访学研修重点项目(gxfx ZD2016190) 安徽大学信息保障技术协同创新中心2015年度开放课题(ADXXBZ201504)
【分类号】：TN92

【相似文献】