CLASSIFYING INTERNET APPLICATIONS ON EARLY STAGE IN 3G ENVIRONMENT

摘要：

Application identification plays a crucial role in network management. Among many different approaches introduced in the field, Bernaille et al. 1 introduces connection based clustering to identify applications. In this paper, first four packet sizes with directions are used as a feature set to cluster different applications in wireless environment, which is acclaimed to cluster with 93.7% of accuracy in 1. However, our experiments not only show that it has low identification accuracy (89.3%), but also show clustering is not suitable for identification of wireless applications.

关键词： Application classification 3G Machine learning Clustering network measurement

作者: Taizhong Quan Seongjin Lee Youjip Won

作者单位: Hanyang university, Korea

会议类型: 国际会议

会议名称: 2009 IEEE International Conference on Network Infrastructure and Digital Content(2009年IEEE网络基础设施与数字内容国际会议 IEEE IC-NIDC2009)

会议地点: 北京

会议语种:英文

页码: 435-439

在线出版日期: 2009-11-06（万方平台首次上网日期，不代表论文的发表时间）