Solving P2P Traffic Identification Problems Via Optimized Support Vector Machines
Solving P2P Traffic Identification Problems Via Optimized Support Vector Machines
复制标题
DOI：
10.1109/aiccsa.2007.370879
复制
发表时间：
2007-05
期刊：
影响因子：
--
通讯作者：
Yue-Xiang Yang;Rui Wang;Yang Liu;Shang-zhen Li;Xiao-yong Zhou
中科院分区：
文献类型：
--
作者：
Yue-Xiang Yang;Rui Wang;Yang Liu;Shang-zhen Li;Xiao-yong Zhou
自上世纪90年代P2P网络出现以来，P2P流量已经成为网络流量中最重要的部分之一。P2P流量的准确识别对于有效的网络管理和网络资源的合理利用具有重要意义。P2P流量的应用级分类，特别是在没有有效负载特征检测的情况下，仍然是一个具有挑战性的问题。提出了一种仅利用传输层信息的P2P流量识别和应用层分类方法。该方法使用的支持向量机已被优化用于执行大型学习任务，从而使该方法变得更适合于大的网络流量。实验结果表明，该方法具有较高的识别效率，适合于实时识别。通过对参数的仔细调整，可以使该方法达到较高的精度。
Since the emergence of peer-to-peer (P2P) networking in the last 90s, P2P traffic has become one of the most significant portions of the network traffic. Accurate identification of P2P traffic makes great sense for efficient network management and reasonable utility of network resources. Application level classification of P2P traffic, especially without payload feature detection, is still a challenging problem. This paper proposes a new method for P2P traffic identification and application level classification, which merely uses transport layer information. The method uses support vector machines which have been optimized for performing large learning tasks, rendering that this method become more suitable for large network traffic. The experimental results show that this method achieved high efficiency and is suitable for real-time identification. And carefully tuning the parameters could make the method achieve high accuracy.