Eecient Transport of Stored Video Using Stream Scheduling and Window-based Traac Envelops

We present eecient scheduling schemes for transporting MPEG-compressed video sources with stringent, deterministic quality-of-service (QoS) requirements. These schemes are used with VBR-coded and archived videos that are transported using a CBR service. A video source is characterized by a time-varying traac envelop, which constitutes an upper bound on the actual bit rate. To provide a relatively tight bound, window-based envelops are used in which the bound is based on xed-length segments of the movie. Using such envelops, we show that video streams can be appropriately scheduled for transmission over the network so that the allocated bandwidth per stream is signiicantly less than the source peak rate (in some cases, 85% reduction in the source peak rate was achieved). This bandwidth gain is obtained via statistical multiplexing, but without sacriicing the stringency and deterministic nature of the QoS. Online procedures for bandwidth computation and admission control under the examined scheduling schemes are introduced, and numerical results are used to demonstrate and compare their performance.

[1]  David K. Y. Yau,et al.  An algorithm for lossless smoothing of MPEG video , 1994, SIGCOMM 1994.

[2]  Satish K. Tripathi,et al.  Impact of video scheduling on bandwidth allocation for multiplexed MPEG streams , 1997, Multimedia Systems.

[3]  Wu-chi Feng,et al.  Smoothing and buffering for delivery of prerecorded compressed video , 1995, Electronic Imaging.

[4]  Martin Vetterli,et al.  Congestion control strategies for packet video , 1991 .

[5]  Kevin C. Almeroth,et al.  The Use of Multicast Delivery to Provide a Scalable and Interactive Video-on-Demand Service , 1996, IEEE J. Sel. Areas Commun..

[6]  Edward W. Knightly,et al.  Fundamental limits and tradeoffs of providing deterministic guarantees to VBR video traffic , 1995, SIGMETRICS '95/PERFORMANCE '95.

[7]  Michael Devetsikiotis,et al.  Modeling and simulation of self-similar variable bit rate compressed video: a unified approach , 1995, SIGCOMM '95.

[8]  Ness B. Shroff,et al.  Video modeling within networks using deterministic smoothing at the source , 1994, Proceedings of INFOCOM '94 Conference on Computer Communications.

[9]  James P. G. Sterbenz,et al.  Networking Requirements for Interactive Video on Demand , 1995, IEEE J. Sel. Areas Commun..

[10]  Keith W. Ross,et al.  Video-on-Demand Over ATM: Constant-Rate Transmission and Transport , 1996, IEEE J. Sel. Areas Commun..

[11]  Dipankar Raychaudhuri,et al.  Bandwidth Renegotiation for VBR Video Over ATM Networks , 1996, IEEE J. Sel. Areas Commun..

[12]  Amy R. Reibman,et al.  Traffic descriptors for VBR video teleconferencing over ATM networks , 1995, TNET.

[13]  Donald F. Towsley,et al.  Supporting stored video: reducing rate variability and end-to-end resource requirements through optimal smoothing , 1996, SIGMETRICS '96.

[14]  Marwan Krunz,et al.  Statistical characteristics and multiplexing of MPEG streams , 1995, Proceedings of INFOCOM'95.

[15]  Ali Tabatabai,et al.  A scheme for smoothing delay-sensitive traffic offered to ATM networks , 1992, [Proceedings] IEEE INFOCOM '92: The Conference on Computer Communications.