基于信令分析的BT流量预识别方法

BitTorrent Traffic Pre-Identification Based on Signaling Analysis

P2P(Peer-to-Peer)系统在内容共享、即时通信、流媒体等领域获得了广泛应用。然而,P2P流量消耗了大部分网络带宽,造成网络拥堵,实现对P2P流量的有效识别和管理成为了一个十分热门的话题。针对BT的信令协议,分析了从Tracker服务器、DHT网络及通过PEX协议,获取节点列表的过程,提出通过相应报文中节点列表的解析,来达到BT流量预识别的方法。该方法能准确预知BT数据流的发生,较准确地对BT流量进行识别,并且有效减少系统开销。

Peer-to-peer（P2P） systems enable a wide range of important,scalable and reliable services such as file sharing,instant messaging（IM） and video streaming.However,excessive P2P traffic can cause significant network management problems because these systems are notoriously greedy with regard to bandwidth consumption.As a result,the effective monitoring and control of BT traffic has become a very hot topic.In this paper,the source of BT peer list is descibed,and then the methods of parsing packets which contain the peer list and pre-identification processing are analyzed.In all,this method realizes an accurate pre-identification of BT traffic which reduces system overhead effectively.