A real-time identification system of unstructured P2P multicast video streaming

The identification of unstructured P2P multicast video streaming is the premise for playing online linkage and real-time evidence in the process of network monitoring management. Based on our preliminary research, a real-time identification system is designed and implemented. The system selects flow features and behavior features which are more real-time and have stronger distinction, adopts the machine learning method of support vector machines, and then successively separates the network traffic until the applications of unstructured P2P multicast video streaming are identified. The system can adapt to the changing network and identify known and unknown applications. Besides, in order to make the network managers recognize and block up the abnormal unstructured P2P multicast video streaming in time, it also has the advantages of strong real-time and low computational complexity.