5G模拟终端信令分析技术研究  

SIGNALING ANALYSIS TECHNOLOGY IN 5G SIMULATOR TERMINAL

作　　者：王宛宛 段浴 Wang Wanwan;Duan Yu(Large Data Institute,Chongqing Vocational College of Transportation,Chongqing 402247,China;School of Communication and Information Engineering,Chongqing University of Posts and Telecommunications,Chongqing 400065,China)

机构地区：[1]重庆交通职业学院大数据学院,重庆402247 [2]重庆邮电大学通信与信息工程学院,重庆400065

出　　处：《计算机应用与软件》2023年第5期148-153,共6页Computer Applications and Software

基　　金：重庆市重点产业共性关键技术创新重大主题专项(cstc2017zdcy-zdzx0030)。

摘　　要：针对5G移动网络数据量暴增、信令分析处理效率较低的问题,为提升信令分析性能,5G与人工智能相结合的智能通信将是移动通信发展的主流方向之一。基于新型5G网络所设计的新型信令分析系统架构,可很好地应用于5G模拟终端。在该架构的基础上,提出一种基于机器学习的信令分析算法。该算法在信令分析过程中,在用户信令流程关联合成之前,选取信令数据Key值C-RNTI作为用户特征。将信令数据中按用户标识进行有监督的信令特征分类模型训练,并使用该模型进行信令分类。实验结果表明,该算法模型综合性能指标较高、分类性能较好,达到了预期效果。Aimed at the issue that the sudden increase of data volume of 5G network and low efficiency of signal processing,in order to improve the performance of signaling analysis,the intelligent communication with the combination of 5G and artificial intelligence will be one of the mainstream directions of wireless communication development.A new signaling analysis system architecture was designed based on the new 5G network,which could be well applied in 5G simulator terminal.On the basis of this architecture,a signaling analysis algorithm based on machine learning was proposed.In the process of signaling analysis,the algorithm selected C-RNTI in key message key value of signaling data as the user feature before the associated synthesis of user signaling process.The algorithm trained the signaling feature classification model according to the user identification C-RNTI under supervised conditions,and used the trained model to perform signaling classification.The experimental results show that the algorithm has higher comprehensive performance index and better classification performance,which has achieved the expected results.

关 键 词：5G模拟终端 信令分析 机器学习 分类训练 用户标识 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术] TN929.5[电子电信—通信与信息系统]