Language Identification of Encrypted VoIP Traffic Alejandra strongystrong.pdfVIP

Language Identification of Encrypted VoIP Traffic: Alejandra y Roberto or Alice and Bob? Charles V. Wright Lucas Ballard Fabian Monrose Gerald M. Masson {cvwright, lucas, fabian, masson}@ Department of Computer Science Johns Hopkins University Baltimore, MD, USA Abstract traffic analysis of encrypted VoIP calls. In this paper, we show that although encryption prevents an eavesdrop- Voice over IP (VoIP) has become a popular protocol for per from reading packet contents and thereby listening in making phone calls over the Internet. Due to the poten- on VoIP conversations (for example, using [21]), traffic tial transit of sensitive conversations over untrusted net- analysis can still be used to infer more information than work infrastructure, it is well understood that the con- expected—namely, the spoken language of the conversa- tents of a VoIP session should be encrypted. However, tion. Identifying the spoken language in VoIP communi- we demonstrate that current cryptographic techniques do cations has several obvious applications, many of which not provide adequate protection when the underlying au- have substantia