基于SDN的细粒度视频流量控制机制研究  

Fine-grained Video Traffic Control Mechanism Based on SDN

作　　者：崔峻玮 翟亚红[1] 周婉旭 贺权 吕勇 Cui Junwei;Zhai Yahong;Zhou Wanxu;He Quan;LüYong(School of Electrical&Information Engineering,Hubei University of Automotive Technology,Shiyan 442002,China)

机构地区：[1]湖北汽车工业学院电气与信息工程学院,湖北十堰442002

出　　处：《湖北汽车工业学院学报》2022年第3期61-65,共5页Journal of Hubei University Of Automotive Technology

基　　金：湖北省教育厅科研计划重点项目(D20211802)。

摘　　要：为了有效地管理网络资源和保障不同多媒体业务的服务质量,设计了基于SDN的细粒度流量控制机制。利用机器学习技术,通过在SDN控制平面中部署卷积神经网络算法,实现了视频流量的细粒度分类,并在SDN实验平台上对该控制机制进行了验证。结果表明:分类准确率达到95%以上,有效实现了视频流差异化传输服务。To effectively manage network resources and guarantee the service quality for different multimedia businesses, a fine-grained traffic control mechanism based on SDN was designed. Through machine learning techniques, the fine-grained classification of video traffic was achieved by deploying convolutional neural network algorithms in the SDN control plane, and the control mechanism was experimentally validated on the SDN experimental platform. The results show that the classification accuracy reaches over 95%, effectively achieving the video stream differentiated transmission service.

关 键 词：软件定义网络 视频流分类 卷积神经网络 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]