Efficient semi-supervised learning bittorrent traffic detection - an extended summary
ICDCN'12 Proceedings of the 13th international conference on Distributed Computing and Networking
Hi-index | 0.00 |
these years, P2P applications are very popular on the Internet and take a big part of the Internet traffic workload. Identifying the P2P traffic and understanding their behavior is an important field. However, previous P2P traffic identification methods by examining user payload or well-defined port numbers no longer adapt to current P2P applications. In this paper, we develop a statistical methodology by analyzing packet length and the P2P basic characteristic, P2P nodes can serve as a server and client at the same time, to identify P2P flows at the transport layer without relying on the port numbers and packet payload.