| With the rapid development of networking social life, the network applications are shifting from traditional businesses based on the C/S and B/S structure to the new network applications based on P2P network structure gradually. A lot of P2P traffic has brought many quality problems to the Internet: bandwidth congestion caused by the surge of network traffic and the network environment deterioration. It’s important to avoid network congestion, limit the illegal traffic and purify network environment by controlling the P2P traffic reasonably. With a further study of the key technical problems: network traffic detection technology and pattern matching algorithms, the paper sets up a protocol signatures library of the traffic monitoring system, and designs and implements the traffic detection module and signatures library maintenance module. The main work of this paper includs the following four aspects:1. Studying the DPI, DFI and deep analytical correlation matching traffic detection methods, and a protocol signatures library is set up by extracting the protocol signatures of30network applications (such as QQ, youku and thunder and so on) with these three methods.2. Design and implement the traffic detection module of the traffic monitoring system. Detect the VoIP traffic with deep analytical correlation matching method. Comparing unknown traffic and the protocol signatures with AC_BM matching algorithm, complete the classification of unknown traffic.3. Design and implement the signatures library maintenance module. The module detects and records whether network applications are upgraded, and the results provide basis with the change of traffic for whether to refresh the signatures library.4、Based on the successful test of the signatures library, traffic detection module and the signatures library maintenance module, the systerm is overall tested and the result shows that this system can control the network traffic effectively with good identification accuracy. |
PDF Full Text Request