计算机工程与应用 ›› 2012, Vol. 48 ›› Issue (25): 98-104.

• 网络、通信、安全 • 上一篇    下一篇

一种基于缓存的SVC流媒体质量自适应算法

黄  威,周  旭,张  棪,白  帆,宋俊平   

  1. 中国科学院 声学研究所 高性能网络实验室,北京 100090
  • 出版日期:2012-09-01 发布日期:2012-08-30

Cache-based quality adaptive algorithm of SVC streaming and its implementation

HUANG Wei, ZHOU Xu, ZHANG Yan, BAI Fan, SONG Junping   

  1. High Performance Network Laboratory, Institute of Acoustics, Chinese Academy of Sciences, Beijing 100090, China
  • Online:2012-09-01 Published:2012-08-30

摘要: 新型视频编码标准H.264/AVC的扩展SVC(可扩展视频编码)技术为未来异构环境下的流媒体传输提供了一种新的解决方案。针对基于SVC的流媒体系统中的核心问题——质量自适应,提出了一种新的算法。该算法分为预缓存和基于缓存的动态质量自适应两个阶段,不但能够自适应网络带宽的动态变化,而且避免了频繁的质量切换,使得用户能够在动态的网络环境下获得尽可能好的主观观看体验。实现了一套SVC流媒体原型系统,并在此基础上验证了算法的有效性。

关键词: 可扩展视频编码(SVC), 质量自适应, 视频传输

Abstract: The extension of Video Coding Standard H.264/AVC—Scalable Video Coding (SVC) provides a promising solution for streaming towards future heterogeneous environment. This paper proposes a new algorithm for the core issue that quality adaptation in streaming systems based on SVC. The algorithm is divided into two stages, pre-buffer and cache-based dynamic quality adaptation, which can not only adapt to network dynamics, but also avoid frequent quality switching, thus enables user to enjoy the best possible viewing experience. Next, it implements a SVC streaming prototype system, and validates the algorithm above in it.

Key words: Scalable Video Coding(SVC), quality adaptation, video transmission