文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > Variable Bit Rate Video in ATM Network Models and Statistical Characteristics
侵权投诉

Variable Bit Rate Video in ATM Network Models and Statistical Characteristics

This report introduces some basic issues regarding the transmission of variable bit rate (VBR) video in asynchronous transfer mode (ATM) networks with concentration on VBR video models and statistical characterises. Performance modeling for VBR video traff

1 Introduction Variable bit rate (VBR) video is predicted to be a major source of tra c for the broadband integrated service digital network (B-ISDN) which is based on asynchronous transfer mode (ATM). The hehavior of ATM networks carrying VBR video tra c has been of interest for a number of years, and has therefore received much attention in the literature. VBR encoders have the potential to provide a constant quality video signal if values of the encoding parameters remain constant or if there is no loss of information due to packet losses. However, VBR video tra c also imposes very stringent real-time constraints on the networks, especially in the case of interactive services. From the network point of view, the inherently complex stream structures of VBR tra c makes the evaluation of the performance of ATM networks di cult under the apperance of VBR services. Thus, analytically tractable models for VBR tra c, which takes into account the tra c charateristics from realistic services that would be carried in an ATM network, and, in particular, which could accurately represent the statistical nature of very bursty, high-speed video services, are required. Statistical characterization and modeling of the output process of VBR video codecs are fundamental steps in understanding how VBR coding a ects network performance. Modeling of VBR coded video tra c deals with the development of a mathematical model by monitoring and studying the output of VBR coded sources. A good model provides a tool to accurately characterize video sources and enables a theoretical assessment of the network performance as well. Two classes of model are needed for the VBR tra c: single source models and multiplexed source models. Single source models could be used to study what types of tra c descriptors make sense for parameter negotiation with the network at the call setup procedure, as well as to test rate control algorithms and predicte the quality of service (QoS). Multiplexed models, in contrast, capture the characteristics of statistically multiplexed bursty

seources and are used to evaluate the aggregated e ect of the superposition of busty streams on network instead of individual source. Video modeling, however, presents some di culties due to the intrinsic complexity and 1

上一页第2页

你可能喜欢

热门文档

相关文档

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email