End-to-End Delay of Videoconferencing over Packet Switched Networks

Videoconferencing is an important global application it enables people around tha globe to interact when they are far from one another. In order for the participants in a videoconfierence call to interact naturally, the end-to-end delay should be below human perception about 100 tus. Since the global propagation delay can be about 100 ms, the actual end-to-end delay budget available to the system designer (excluding propagation delay) can be no more than 10 rns. We identifi the components of the end-to-end delay in various couiigurations with the object ive of understanding how it can be kapt below the desired 10 ms bound. We analyze these components going step-by-step through six system configurate ions obtained by comkdning three generic network nrchltect ures with two video encoding schemes. We study the transmission of raw video and variable tdt rate (VBR) MPEG video encoding over (i) circuit switching, (ii) synchronous packet switching, and ( ;ii) rrsynchronous packet switching. In addition, we show that constant bit rate (CBR) MPEG encoding delivers unacceptable delay, which is on the order of the group of pictures (GOP) time interval. ‘I%is ~t,~dy ShOWs that having a global common time. refamnc% together with time-driven priority (TDP) and VBR MPEG video encoding, provides adequate end-to-end delay, which is (i) below 10 ms, (ii) independent of the network instant load, and (iii) independent of the connection rate. The rssulting end-to-end delay (excluding propagation del~) can be smaller than the video frame period, wKlch is better than what can be obtained with circuit switching. KeywordsTime-driven Priority, MPEG, Videocrmference, Quality of Service End-to-end Delay, Performance Guarantees.