Skip to Main Content
The convergence of voice and data on Internet Protocol (IP) networks has led to tremendous growth of Voice over IP (VoIP). While SIP/ H.323 are dominant signaling protocols, RTP is widely used for media transfer. Certain situations may require logging of RTP sessions along with the signaling information for call and voice analysis. Recovering voice from RTP packets requires explicit handling of RTP stack and decoding. In this paper, we propose a scheme that uses softphones with appropriate codec support for voice recovery from logged RTP packets. It is done using inherent provisions of RTP, specifically SSRC-multiplexing. The scheme also suggests the possibility of an active stream injection attack in RTP based VoIP.
Date of Conference: 12-13 Dec. 2011