Optimized streaming of paketized video

Eren Gurses, G. Bozdagi Akar, Nail Akar (Bilkent Univ.)
 

PERFORMANCE OF H.263+ SCALABLE VIDEO OVER A DIFFSERV NETWORK

The transmission of high quality video over the Internet is now becoming a reality due to progresses in video compression, networking technologies, efficient video coders/decoders andincreasing interest in applications such as video on demand, videophone, and videoconferencing.  To fulfill different requirements by using one common bitstream in a wide range of video services in heterogenous environments, techniques which can simultaneously support a variety of bitrates tailored to individual services are needed while maintaining end-to-end quality. Coding video in a scalable manner partially solves this problem by offering different rates to different users. For maintaining end-to-end quality, two QoS (Quality of Service) architectures have been proposed by the IETF (Internet Engineering Task Force): the integrated services (IntServ) with the resource reservation protocol (RSVP) and the differentiated services (Diffserv). Diffserv provides a less complicated and scalable solution compared to Intserv, which fits very well to the structure of scalable video coding. Recently, several studies have been done on transmitting scalable video (MPEG-2, H.263+, MPEG-4) over Diffserv networks. Markopoulou and Hang address the issue of transmission of scalable video (H.263+) in contexts where packet drops, rather than packet delays, are the primary determinant of application performance. However, in this work only SNR scalability is used and there is no policing algorithm involved at the edge to check the conformance of incoming packets. Shin et. al. use a relative priority index to represent the relative preference of each packet in terms of loss and delay. Instead of using scalable video, their work is based on full scale video. Ahmed et. al. investigate the transmission of MPEG-4 bitstreams over a DiffServ network. Their work mostly concentrates on the encapsulation of MPEG-4 packets over RTP/IP and a marking mechanism at the Diffserv edge routers.
In this work, we evaluate the performance of transmitting H.263+ scalable video over a Diffserv network using different SNR and temporal scalabilities, in a realistic scenario in which policing is done at the edge of a Diffserv domain to check conformance. Our simulation results demonstrate that if the interframes selected by the reference picture selection mode of H.263+ are transmitted in the base layer (rather than the enhancement layer), better bandwidth utilization and error resilience is achieved in comparison with the following three cases: i) non-scalable coding ii) SNR scalability iii) temporal scalability without reference picture selection. We also introduce a promotion/demotion policer by which packets are promoted/demoted at the edge of a Diffserv  domain according to a novel policing algorithm we propose in this paper.