基于高铁沿线用户分类的LTE专网优化  

Optimization of LTE Private Networks Based on High-speed Rail User Classification

作　　者：Alexis Huet 徐珊珊 王计斌 金栋梁[2] 许正[2] Alexis Hue;XU Shanshan;WANG Jibin;JIN Dongliang;XU Zheng(Nanjing HOWSO Technology,Nanjing 210000,China;College of Automation,Nanjing University of Posts and Telecommunications,Nanjing 210000,China)

机构地区：[1]南京华苏科技有限公司,江苏南京210000 [2]南京邮电大学自动化学院,江苏南京210000

出　　处：《移动通信》2018年第12期86-91,共6页Mobile Communications

摘　　要：通过机器学习中的随机森林算法对高速铁路沿线用户进行分类识别,从训练数据集中分离出专网用户和大网用户,且对训练集和测试集进行精度测试。然后,对LTE小区连接情况进行分析以达到评估高速铁路整体网络线路的目的。最后,对网络质量差的小区进行优化。The users along high-speed rail are classified by using the random forest algorithm in machine learning so that the users of private networks and the users of common networks are separated from training data set. The accuracy test is performed for the training set and testing set. Then, the connection situation of LTE cells is analyzed to evaluate the entire high-speed rail network. Finally, the network optimization is performed for the bad-quality cells.

关 键 词：机器学习 高速铁路 LTE 网络优化 

分 类 号：TN929.5[电子电信—通信与信息系统]