Classification of Audio Codecs in Telecommunication Networks

Audio data are widely exchanged in telecommunications networks. Due to the limitation of network resources, these data are usually compressed before transmission. Various methods exist for compressing audio data. Hence, in order to unlawfully access these audio information, one needs to first identify the codec which is used for audio compression. One of the best approaches for audio codec identification is identification based on the contents of received packets. In these methods, statistical features of received packets are used for identification of employed codec. In this paper, a method of audio codec classification is proposed based on longest common substring and longest common subsequence features. Simulation results (with accuracy of 97% for 8 Kbytes packets) demonstrate the superiority of the proposed method compared to conventional methods.