信号域距离对室内WKNN指纹定位精度的影响分析  

Analysis of the influence of signal-domain distance on the indoor WKNN fingerprint positioning accuracy

作　　者：毕京学 纪冬华[2] 王志华 李亚宇 刘晓[1] 李秋琳 BI Jingxue;JI Donghua;WANG Zhihua;LI Yayu;LIU Xiao;LI Qiulin(Shandong Jianzhu University,Jinan 250101,China;Provincial Geomatics Center of Jiangsu,Nanjing 210013,China)

机构地区：[1]山东建筑大学,济南250101 [2]江苏省基础地理信息中心,南京210013

出　　处：《导航定位学报》2020年第5期37-41,共5页Journal of Navigation and Positioning

基　　金：国家重点研发计划项目(2016YFB0502102);卫星导航系统与装备技术国家重点实验室开放基金项目(CEPNT-2018KF-03);山东省高等学校青创人才引育计划项目;山东建筑大学开放实验项目(2019yzkf159)。

摘　　要：室内加权K近邻(WKNN)指纹定位的关键是测试指纹与参考指纹间信号域距离的表达,现有研究对信号域距离计算的研究不够深入。为研究信号域距离对室内指纹定位的影响,根据遍历指纹以及求和与求均值计算对信号域距离进行分类,利用WKNN算法在3200 m2的试验场基于不同信号域距离开展指纹定位测试。基于遍历测试指纹计算得到信号域距离的定位平均误差(ME)为2.398 m,均方根误差(RMSE)为1.903 m。与其他信号域距离相比,ME减小了至少0.648 m,RMSE减小了至少0.573 m。实验结果表明:信号域距离对指纹定位精度的影响较大,基于遍历测试指纹计算得到信号域距离的指纹定位精度最高。The key of indoor fingerprinting localization based on weighted K nearest neighbors(WKNN)is the expression of the distance between testing fingerprint and reference fingerprint.The existing researches on signal-domain distances are lack of thorough study.In order to study the influence of signal-domain distances on indoor fingerprinting localization,classification of signal-domain distances were provided,according to traversing fingerprint,sum and mean calculation.Fingerprinting localization tests were conducted based on several different signal-domain distances by using WKNN algorithm in a testbed with 3200 m2,separately.The fingerprinting localization based on signal-domain distance,calculated by traversing testing fingerprint,achieves the mean error(ME)of 2.398 m,the root mean square error(RMSE)of 1.903 m.Compared with other signal-domain distances,the ME decreases by more than 0.648 m,the RMSE decreases by more than 0.573 m.Experimental results show that it has a good performance for fingerprinting localization based on signal-domain distance by traversing testing fingerprint.

关 键 词：室内定位 指纹定位 信号域距离 加权K近邻 接收信号强度 

分 类 号：TN914[电子电信—通信与信息系统]