一种基于自主学习的DPI业务识别方法
发布时间：2018-09-18 14:08
【摘要】：随着数据通信技术的发展,因特网服务提供商面临着沉重的P2P和Vo IP流量管理问题.针对目前已有的业务识别方法大都存在带宽消耗价大,实现复杂等问题,提出一种基于自主学习的深度包检测(DPI)业务识别方法,利用关联流量和统计分析等方法识别新的业务流量.能够在统计系统中实时的增加新的识别流矢量,因此无需人工分析和标记就能够对统计数据集进行更新.实验结果证明了该方法的准确性和流量识别的高度可靠性.
[Abstract]:With the development of data communication technology, Internet service providers are faced with heavy P2P and Vo IP traffic management problems. Aiming at the problems of high bandwidth consumption and complex implementation in most of the existing service identification methods, a new (DPI) service identification method based on autonomous learning is proposed. New traffic is identified by association traffic and statistical analysis. A new recognition stream vector can be added to the statistical system in real time, so the statistical data set can be updated without manual analysis and marking. The experimental results show the accuracy of the method and the high reliability of flow identification.
【作者单位】： 福建师范大学现代教育技术中心;福建师范大学数学与计算机科学学院;
【分类号】：TP393.06
本文编号：2248154
[Abstract]:With the development of data communication technology, Internet service providers are faced with heavy P2P and Vo IP traffic management problems. Aiming at the problems of high bandwidth consumption and complex implementation in most of the existing service identification methods, a new (DPI) service identification method based on autonomous learning is proposed. New traffic is identified by association traffic and statistical analysis. A new recognition stream vector can be added to the statistical system in real time, so the statistical data set can be updated without manual analysis and marking. The experimental results show the accuracy of the method and the high reliability of flow identification.
【作者单位】： 福建师范大学现代教育技术中心;福建师范大学数学与计算机科学学院;
【分类号】：TP393.06
【相似文献】
相关期刊论文 前5条
1 刘梦莲;;基于网络的口译自主学习支持应用研究[J];电脑与电信;2013年07期
2 龙虎;李娜;;基于虚拟社区的交互式自主学习系统设计探究[J];电脑知识与技术;2013年18期
3 姜少红;;计算机网络教学环境下学生自主学习能力的培养[J];中国成人教育;2014年12期
4 张慧丽;;Moodle平台自主学习的设计研究——以“HTML语言基础”主题内容为例[J];软件导刊(教育技术);2013年11期
5 ;[J];;年期
，本文编号：2248154
本文链接：https://www.wllwen.com/guanlilunwen/ydhl/2248154.html
