基于VBR视频流量预测的动态带宽分配滞回算法

刘亚伟; 荆涛; 蒋笑冰; 冯玉珉

基于VBR视频流量预测的动态带宽分配滞回算法

计量
- 文章访问数: 2483
- HTML全文浏览量: 108
- PDF下载量: 735
- 被引次数: 0
出版历程
- 收稿日期: 2003-12-25
- 修回日期: 2004-04-19
- 刊出日期: 2005-05-19

A Hysteresis Algorithm for Dynamic Bandwidth Allocation Based on VBR Video Traffic Prediction

摘要

摘要: 研究表明VBR视频流量具有较强的长时相关性和突发性,该文利用高阶自回归过程预测其流量,并进行概率分布转换,同时提出了一种新的动态带宽分配算法--滞回算法,与Chiruvolu(1998)提出的门限算法,及Iraqi(1999)提出的基于GOP场景检测的带宽分配方案相比,对缓存尺寸的需求、信元丢失率等性能有明显提高。仿真表明,该算法能够以较低的开销(带宽重分配次数百分比2%)同时兼顾了信元丢失率和带宽利用率,而且只有在缓存队长超过迂回门限时才触发带宽重分配过程,有利于在线实现。
- VBR视频流量预测; 尾部概率分布; 动态带宽分配; 滞回算法; 信元丢失率
Abstract: With lots of studies, it is admitted that the variable bit rate video traffic exhibits high long range dependence and burstness. Its traffic is predicted by auto-regressive process with high order in this paper while the probablity distribution of traffic predicted is also transformed to that of real situation. And then a new dynamic bandwidth allocation algorithm named hysteresis algorithm is presented, by which performances of buffer size requirement and cell loss rate are both improved significantly compared with threshold algorithm presented by Chiruvolu (1998) and allocation scheme based on group of pirclures scene detection presented by Iraqi (1999). The proposed algorithm gives attention to both cell loss rate and utilization at the considerate low cost (reallocation frequency ^2%) according to a number of simulations. And the bandwidth reallocation process is only triggered when the queuing length of buffer surpasses the roundabout threshold so that it is easy for online realization.

HTML全文

参考文献(1)

Sivaradje G, Dananjayan P. Effcient resource allocation scheme for real-time MPEG video traffic over ATM networks. The 8th International Conference on Communication Systems, Nov. 2002,Vol.2:747 - 751.[2]Iraqi Y, Boutaba R. A dynamic bandwidth allocation algorithm for MPEG video sources in wireless networks. Proceedings of the 3rd International Workshop on Discrete Algorithms and Methods for Mobile Computing and Communications, Seattle, Washington,United States, August 1999:86 - 92.[3]Lei H, Kumar S, Kuo C.-C. J. Adaptive resource allocation for multimedia services in wireless communication networks. 2001International Conference on Distributed Computing Systems Workshop, April 2001: 307 - 312.[4]Nagarajan K, Zhou G T. A new resource allocation scheme for gaussian traffic sources. 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing, June 2000, Vol.5:2609 - 2612.[5]Chiruvolu G, Sankar R, Ranganathan N. Adaptive VBR video traffic management for higher utilization of ATM networks[J].ACM SIGCOMM Computer Communication Review.1998, 28(3):27-[6]Supangkat S H, Nugraha I G B B. Characteristic of variable- bitrate video frames duration over the ethernet network. Fifth AsiaPacific Conference on Communications and Fourth Optoelectronics and Communications Conference, October 1999,Vol. 1: 94 - 97.[7]Doulamis N D, Doulamis A D, Konstantoulakis G E, et al..Efficient modeling of VBR MPEG-1 coded video sources[J].IEEE Trans. on Circuits and Systems for Video Technology.2000, 10(1):93-[8]Garrett M W, Willinger W. Analysis, modeling and generation of self-similar VBR video traffic. Proc. ACM SIGCOMM94,London, U. K., August 1994: 269 - 280.[9]Krunz M M, Makowski A M. Modeling video traffic using M/G/ input process: a compromise between markovian and LRD models[J].IEEE J. on Selected Areas in Communications.1998, 16(5):733-[10]Liu D, Sra E I, Sun W. Nested auto-regressive progress for MPEG-encoded video traffic modeling[J].IEEE Trans. on Circuits and Systems for Video Technology.2001, 11 (2):169-

施引文献

资源附件(0)

访问统计