Wide area traffic: the failure of Poisson modeling
IEEE/ACM Transactions on Networking (TON)
HMM profiles for network traffic classification
Proceedings of the 2004 ACM workshop on Visualization and data mining for computer security
Internet traffic classification using bayesian analysis techniques
SIGMETRICS '05 Proceedings of the 2005 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
BLINC: multilevel traffic classification in the dark
Proceedings of the 2005 conference on Applications, technologies, architectures, and protocols for computer communications
Automated Traffic Classification and Application Identification using Machine Learning
LCN '05 Proceedings of the The IEEE Conference on Local Computer Networks 30th Anniversary
Traffic classification on the fly
ACM SIGCOMM Computer Communication Review
Traffic classification through simple statistical fingerprinting
ACM SIGCOMM Computer Communication Review
A Machine Learning Approach for Efficient Traffic Classification
MASCOTS '07 Proceedings of the 2007 15th International Symposium on Modeling, Analysis, and Simulation of Computer and Telecommunication Systems
Internet traffic classification demystified: myths, caveats, and the best practices
CoNEXT '08 Proceedings of the 2008 ACM CoNEXT Conference
| Hi-index | 0.00 |
Accurate traffic classification is a necessary means of network management, QOS, monitoring and so on. We find that each protocol's flows have their own packet-level rhythm on the statistical characteristics. In this paper we present a Bayesian network classification mechanism based on the flows' packet-level rhythm. However, the flows rhythm is always too scattered to bring into play its ability well in the Bayesian network, so we employ an Equal-width discretization method to centralize the rhythm and discretize the packet size and interval-time to some different space. Then we applied our classification model to the different discretization data set of HTTP, EDONKEY, BITTORRENT, FTP and AIM. Experiment results show that our approach can achieve better precision and recall rate for these applications.