5G网络环境下的网络视频质量分析  

Network Video Quality Analysis under the Condition of 5G Network

作　　者：罗锐 黄鹂声[2] 张振宇 薛红 欧阳千玮 谈澄秋 LUO Rui;HUANG Li-sheng;ZHANG Zhen-yu;XUE Hong;OUYANG Qian-wei;TAN Cheng-qiu(School of Computer Science,Chengdu Normal University,Chengdu611130,China;School of Computer Science and Engineering,University of Electronic Science and Technology of China,Chengdu611731,China;Key Lab of Information Coding and Transmission,Southwest Jiaotong University,Chengdu611756,China;International College,Chongqing University of Posts and Telecommunications,Chongqing400065,China;China Mobile Group Sichuan Company Limited,Chengdu610095,China)

机构地区：[1]成都师范学院计算机科学学院,成都611130 [2]电子科技大学计算机科学与工程学院,成都611731 [3]西南交通大学信息编码与传输重点实验室,成都611756 [4]重庆邮电大学国际学院,重庆400065 [5]中国移动通信集团四川有限公司,成都610095

出　　处：《兰州交通大学学报》2022年第6期54-61,共8页Journal of Lanzhou Jiaotong University

基　　金：四川省科技计划资助(2021YFG0169);成都师范学院科技项目(111/111159001);。

摘　　要：网络视频业务是5G时代无线流量的主要增长点,网络视频业务质量监测对于通信运营商提前发现问题、提前解决问题和提高市场竞争力至关重要.为此,根据5G网络大容量、时变衰落无线信道特点和网络视频播放的缓冲机制,在基于通信运营商呼叫/事务/会话数据记录大数据,抽取、加工生成反映视频播放质量的特征信息指标过程中,引入下行丢包率集中度指标,在此基础上,应用机器学习算法,计算得到5G网络环境下的网络视频质量评分值.实验结果表明:该方法在充分利用通信运营商现有统一深度包检测系统、节约成本的同时,由于在抽取、加工生成特征信息指标的过程中引入了下行丢包率集中度指标以反映5G信道变化对视频播放质量的影响,与现有方法相比,视频质量分析具有更高的准确性.Network video service is the main growth point of wireless traffic in 5 G era.Quality of service monitoring of network video service is very important for carriers to detect and solve problems in advance and improve their market competitiveness.Therefore,according to large capacity and time-varying fading wireless channel characteristics of 5 G network and network video play buffer mechanism,in the procedure of extracting,processing and producing a set of characteristic information metrics reflecting the quality of network video playing based on extend data recording(XDR)of call,transaction,session data of communication operators,downlink packet loss rate concentration degree metrics is introduced.On this basis,the network video quality score of 5 G network environment is calculated based on machine learning algorithm.Experimental results show that,due to the introduction of downlink packet loss rate concentration degree metrics which reflect the effect of variation of 5 G wireless channel state on video playing quality,the proposed method not only saves cost by making full use of existing deep packet inspection(DPI)system of carrier,but also has higher accuracy compared with existing network video quality analysis method.

关 键 词：5G 网络视频 呼叫/事务/会话数据记录 质量 下行丢包率集中度 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]