基于大数据思维的MDT数据过滤与室内识别方法  

MDT Data Filtering and Indoor Identification Methods Based on Big Data Thinking

作　　者：郭向荣[1] 李刚[1] 纪纯妹[1] 陈炯锋 马于飞 熊雄 GUO Xiangrong;LI Gang;JI Chunmei;CHEN Jiongfeng;MA Yufei;XIONG Xiong(China Mobile Group Guangdong Co.,Ltd.,Shantou Branch,Shantou 515041,China;China Mobile Group Guangdong Co.,Ltd.,Guangzhou 510000,China)

机构地区：[1]中国移动通信集团广东有限公司汕头分公司

出　　处：《移动通信》2019年第7期91-96,共6页Mobile Communications

摘　　要：现阶段运营商的室内网络评估方法不能有效支撑海量的室内网络评估与分析应用。从MDT数据特性出发,研究MDT数据结构与字段,采用基于大数据多维度分析的MDT异常GPS过滤方法和基于大数据匹配的MDT室内数据识别方法,该方法经过验证能够有效输出室内定位数据,解决MDT定位数据无法用于室内网络评估的问题,提供海量有效经纬度数据用于室内评估与分析。The current indoor network evaluation approaches owned by operators can not effectively support massive indoor network evaluations and analytical applications. The MDT data structures and fields are investigated from the characteristics of the MDT data via two main methods, namely, the MDT abnormal GPS filtering based on the big data multi-dimensional analysis and the MDT indoor data identifi cation based on the big data matching. Both methods are demonstrated to be effective to output indoor positioning data, solve the problem that MDT positioning data can not be used in indoor network evaluations, and provide valid data for massive indoor evaluations and analytical applications.

关 键 词：MDT定位 大数据思维 数据过滤 室内识别 

分 类 号：TN929.5[电子电信—通信与信息系统]