A smooth broadcasting scheme for VBR-encoded hot videos

  • Authors:
  • Hsiang-Fu Yu;Hung-Chang Yang;Pin-Han Ho;Yi-Ming Chen;Li-Ming Tseng

  • Affiliations:
  • Computer Center, National Central University, Taiwan;Department of Computer Science and Information Engineering, National Central University, Taiwan;Department of Electrical and Computer Engineering, University of Waterloo, Canada;Department of Information Management, National Central University, Taiwan;Department of Computer Science and Information Engineering, National Central University, Taiwan

  • Venue:
  • Computer Communications
  • Year:
  • 2006

Quantified Score

Hi-index 0.24

Visualization

Abstract

One way of broadcasting a popular video is partitioning the video into segments, each being broadcasted on several streams periodically. The approach alleviates the bandwidth thirsty in real-time video broadcasting without sacrificing viewers' waiting time by allowing multiple users to share the same video segments. One representative scheme of the category is the recursive frequency-splitting (RFS) broadcasting scheme, which yields approximate minimum waiting time. In this paper, a novel approach is introduced to enhance RFS, called Smooth RFS (SRFS), in which the approaches of segment patching and asynchronous downloading-playing are proposed for achieving a smooth broadcast of variable-bit-rate (VBR) -encoded videos. The design of SRFS aims to reduce the peak bandwidth consumption and variance during the distribution of VBR videos in order to achieve better upper bounds on the bandwidth peak and the variance. Extensive simulation has been conducted on the proposed scheme by comparing a number of past reported counterparts, including the trace-adaptive fragmentation (TAF) scheme and the smooth fast broadcasting (SFB) scheme. The results indicate that the proposed scheme yields lower bandwidth peak and variance. Besides, given a fixed bandwidth that is smaller than the peak, SRFS obtains a much smaller blocking rate than that of the other schemes.