Characterizing packet audio streams from internet multimedia applications.pdf

Characterizing Packet Audio Streams from Internet Multimedia Applications Chen-Nee Chuah Department of Electrical and Computer Engineering University of California, Davis Davis, CA 95616, USA Randy H. Katz Department of Electrical Engineering and Computer Sciences University of California, Berkeley Berkeley, CA 94720, USA Abstract— We analyzed 70 voice traces collected from IP- telephony applications, multicast lectures, and multimedia con- ferencing sessions which involve multiple speakers and different dynamics of interaction beyond two-way conversations. Results show that application differences have significant impact on the traffic characteristics. The conventional exponential model, estab- lished for telephone conversations, fails to accurately capture the packet level activity observed in these traces, e.g., the heavy-tail distributions of the talkspurt and silence periods. We classify the traces into four types based on their audio contents: audience, lec- ture, multi-party conferencing and conversation. Further analy- sis shows that Weibull is a better matching statistical model and achieves lower mean-square-error than the exponential model (by 1 to 2 orders of magnitude) in approximating the audio streams for all four cases. I. INTRODUCTION The startling growth of Internet access, and advance in IP- technology have resulted in a push for an integrated “IP-based core” architecture to support data, voice and wireless access. This has driven the computer-telephony integration, and large- scale deployment of Voice over IP (VoIP) is expected in the near future. Packet audio refers to real-time transmission of digitized audio information over the Internet, and is commonly used in desk-top conferencing multimedia applications such as Netmeeting, NeVot, vic vat and CuSeeMe. With the ubiqui- tous streaming audio players and advance in new technologies like speech recognition, packet voice begins to constitute an in- creasing fraction of Internet traffic. The majo