基于大数据对运营商基站覆盖中心点的预测及对比分析  

Prediction and Comparative Analysis on Coverage Center of the Operator' Base Station Based on Big Data

作　　者：唐忠林 许盛宏[1] 谭志远[1] 

机构地区：[1]中国电信股份有限公司广东研究院,广东广州510630

出　　处：《移动通信》2017年第22期1-4,共4页Mobile Communications

摘　　要：为深入了解运营商的网络覆盖程度,提升网络资源投放效率,通过Mean-Shift算法对基站的MR数据做首次聚类分析,预测出局部最优的基站覆盖中心点,再用DBSCAN算法预测出全局最优的基站覆盖中心点。在此基础上分析三大运营商基站小区在地理位置上的部署密集程度,从而获得每个运营商的网络热点区域分布,为网络规划和智慧网优提供全方位的分析方法。In order to deeply acquaint operators＇ network coverage level and enhance the deployment of network resources, the first clustering analysis on MR data was done based on Mean-Shift algorithm to predict the locally optimal center of the base station coverage. Then, DBSCAN algorithm was used to predict the globally optimal center of the base station coverage. Based on this, the geographic deployment density of base stations for the major three operators was analyzed to obtain the network hotspot area distribution of each operator. It provides a comprehensive analysis method to the network planning and intelligent network optimization.

关 键 词：MR MEAN-SHIFT DBSCAN 聚类算法 

分 类 号：TP312[自动化与计算机技术—计算机软件与理论]