Online hybrid internet traffic classification algorithm based on signature statistical and port methods to identify internet applications

Internet traffic classification gained significant attention in the last few years. Most of the current classification methods were only valid for offline classification. Each of the three common classification methods (port, payload, statistics) has some limitations. To increase the value of Internet traffic classifiers, this paper combines the three methods to produce a new classification algorithm (SSPC). In the proposed algorithm, each traffic flow was classified in parallel three times by one of the three method classifiers. Based on certain priority rules, SSPC makes classification decisions for each traffic flow. The SSPC algorithm was tested by classifying WWW applications traffic in two stages: offline and online. The results of both cases show that SSPC is the higher accuracy when compared with other classifiers. In addition, the results indicate that the SSPC algorithm was suitable for online classification decisions, which is taken with the speed of the Inter-net traffic.