Automatic Application Signature Extracting Method Based on Improved LCS
Sheng He, Junyong Luo, Yan Liu, Xiaoxiao Ren, Jin Zhang · International Conference on Multimedia Information Networking and Security · 2013
Traditionally, network traffic classification is based on well-known port numbers. However, this method is turning more and more inaccurate with the development of the Internet. Lately, the method using application signatures is becoming increasingly popular and it is more accurate than the traditional one based on port numbers. This paper proposes an automatic method for application signature extraction based on improved LCS algorithm. To evaluate the method proposed, signatures are extracted from five common protocols, including HTTP, DNS, OICQ, Kugoo and BitTorrent. The experiment result shows that the method is able to automatically and effectively extract precise, integrated and reliable application signatures with extremely high accuracy.