End-to-end delay analysis of videoconferencing over packet-switched networks

Videoconferencing is an important global application-it enables people arou nd the globe to interact when distance separates them. In order for the par ticipants in a videoconference call to interact naturally, the end-to-end d elay should he below human perception; even though an objective and unique figure cannot be set, 100 ms is widely recognized as the desired one-way de lay requirement for interaction. Since the global propagation delay can be about 100 ms, the actual end-to-end delay budget available to the system de signer (excluding propagation delay) can be Ilo more than IO ms. We identif y the components of the end-to-end delay in various configurations with the objective of understanding how it can be kept below the desired IO-ms boun d. We analyze these components step-by-step through six system configurations obtained by combining three generic network architectures with two video en coding schemes. We study the transmission of ra,tr video and variable bit r ate (VBR) MPEG video encoding over 1) circuit switching; 2) synchronous pac ket switching; and 3) asynchronous packet switching. In addition, we show t hat constant bit rate (CBR) MPEG encoding delivers unacceptable delay-on th e order of the group of pictures (GOP) time interval-when maximizing qualit y for static scenes. This study aims at showing that having a global common time reference, toge ther with time-driven priority (TDP) and VER MPEG video encoding, provides adequate end-to-end delay, which is 1) below 10 ms; 2) independent of the n etwork instant load; and 3) independent of the connection rate. The resulti ng end-to-end delay (excluding propagation delay) can be smaller than the v ideo frame period, which is better than what can be obtained with circuit s witching.