室内指纹定位算法中的Radio.docx

室内指纹定位算法中的Radio?map重构技术李欣欣，邓中亮摘要：针对室内定位中的指纹定位算法的精度和响应时间严重依赖于指纹地图(Radio-map)， 而Radio map 的建立与更新慢，随着定位区域的增大其数据量也不断增大，限制该算法普及 的重要原因，针对这个问题本文提出了Radio-map 重构技术。通过对指纹地图进行异常数据 的去除，结合传播模型对radio-map 进行区域的划分，在区域内进行AP 选择和降维，最后 采用WKNN 算法进行定位匹配，在保证定位精度的情况下减少了定位运算的时间和数据库的 10 容量。在室内环境下对该定位算法进行了实验，结果表明该方法使得指纹地图的数据采集量 大大减少，定位响应时间有明显的提高，定位精度有一定的提高，对于室内定位的普及具有 参考价值。 关键词：室内定位；传播模型；Radio map 重构；WKNN 算法Radio map reconstructed technique in indoor Fingerprint positioning algorithmLI Xinxin, DENG Zhongliang(Beijing University of Posts and Telecommunication, Beijing 100876)Abstract: The fingerprint for the positioning of indoor positioning accuracy and response time algorithms rely heavily on fingerprint map (Radio-map), while the radio map creation and update slowly, with the positioning of the region increases the data volume is increasing, and limit the algorithm is an important reason for popularity, this paper presents for Radio-map reconstruction technique. Fingerprint map through the removal of abnormal data, combined with propagation 25 models for radio-map of regional division, in the region for AP selection and dimensionality reduction, and finally locate the matching algorithm using WKNN, to ensure positioning accuracy in the case of reduced positioning time and computing capacity of the database. In the indoor environment of the location algorithm for the experimental results show that the method makes fingerprint map data acquisition is greatly reduced, positioning significantly improved response 30 time, positioning accuracy is improved to some extent, for the popularity of indoor positioning reference value. Keywords: Indoor positioning; Propagation model; Radio map reconstruction; WKNN algorithm 0引言随着移动设备及无线局域网的普及，基于移动终端位置的服务 [1] (Location-based Services，LBS)受到越来越多的关注。目前，在室外环境下如运用比较广泛的全球定位系统 GPS，在许多领域得到应用，然而在建筑物或室内环境中，楼宇等建筑物阻隔了卫星信号的 传播或者将它们分散开去，造成 GPS 系统定位误差大或者无法定位，即所谓的“都市峡谷” 效应。在室内定位研究领域，有基于到达时间、基于到达角度、基于接收信号强度几种技术。由于基于接收信号强度的指纹识别(Fingerprinting)的定位技术使用范围广，软硬件实现成本 低，成为位置感知领域