基于DiffServ网络的流视频分发系统优化
158 浏览量
更新于2024-08-27
收藏 739KB PDF 举报
本文档主要探讨了一种面向DiffServ (DS)网络的流视频传输系统。DiffServ是一种基于策略的网络服务模型，它通过将流量划分为不同的优先级，以适应不同类型的应用需求并提供可预测的服务质量(QoS)。在这个名为"Delivery System for Streaming Video Over DiffServ Networks"的研究中，作者Lishui Chen和Guizhong Liu提出了一种创新的解决方案。
在该系统中，应用层的编码器根据当前的配置和编码特性，对压缩的数据进行预标记，这种预标记体现了数据的重要性。这种预标记考虑了视频流的不同部分可能对实时性、清晰度等服务质量参数的需求差异。预标记的目的是为了确保关键的视频内容能够获得更高的带宽保障。
预标记被映射到实时传输协议层的固定数量的源标记上，这一步骤有助于在多条路径上传输时，路由器能够理解和处理这些差异化服务代码点(DSCP)。边缘路由器根据网络当前的拥塞状态，利用改进的两速率三颜色标记器(Improved Two Rate Three Color Marker, ITRTCM)技术，动态地调整源标记，从而更有效地分配带宽资源。
核心路由器则根据每跳转发行为来传输这些带有源标记的分组，确保了在多节点传输过程中，高优先级的视频流能够优先处理，避免了可能的丢包或延迟，从而显著提升了流媒体应用的端到端交付质量。
实验结果显示，这种预标记和映射机制能够在DS网络环境中显著提高视频流的观看体验，对于实时性和稳定性要求高的视频服务来说，具有很高的实用价值。因此，本文的研究不仅为DiffServ网络中的视频流媒体传输提供了一个有效的优化方案，还展示了如何通过智能预标记和QoS管理来提升多媒体服务的质量。
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, VOL. 20, NO. 9, SEPTEMBER 2010 1255
Transactions Letters
A Delivery System for Streaming Video Over DiffServ Networks
Lishui Chen and Guizhong Liu
Abstract—In this letter, we develop a general delivery system
for streaming using various coding standards over DiffServ (DS)
networks. In the delivery system, the coder at the application
layer pre-marks the compressed data according to their impor-
tance in terms of the current configuration and characteristics
of the coder. Then the system maps these pre-marks to a fixed
number of source marks at the real-time transport protocol layer.
In terms of the current network congestion state, the edge router
of a DS network uses the improved two rate three color maker
to map these source marks to differentiated services code points.
At last, the core routers transmit these packets according to the
per-hop forwarding behaviors. Experimental results show that
this system can greatly improve the end-to-end delivery quality
of a streaming application.
Index Terms—DiffServ (DS) networks, improved two rate three
color maker (ITRTCM), pre-marking, QoS, two markers system.
I. Introduction
With the rapid development of multimedia applications over
the Internet, such as distant learning, video conference, voice
over Internet protocol, video on demand (VoD), and so on,
new broadband technologies have been developed extensively
in these years. But with the fast increase of network users,
especially the multimedia users, the issue of bandwidth lack is
still a major problem in streaming applications. So, the Internet
Engineering Task Force has been proposed the integrated
services (IntServ) [1] and the differentiated services (DiffServ)
[2] to provide end-to-end quality of service (QoS). DiffServ
(DS) network is more attractive for its simplicity and scalabil-
ity because it never provides signaled QoS for individual flows
as IntServ, but provides static QoS to traffic aggregation.
In order to use the DS architecture in a better way, a
lot of researchers have proposed media-aware application-
level marking schemes, which mark the packets according
to their importance measured in terms of some special rules
and map these source marks to differentiated services code
Manuscript received August 12, 2008; revised January 15, 2009 and July
31, 2009. Date of publication July 12, 2010; date of current version September
9, 2010. This work was supported in part by the National 863 Program
under Project 2009AA01Z409, by the National 973 Project under Project
2007CB311002, and by the National Natural Science Foundation of China
(NSFC) under Project 60972066. This paper was recommended by Associate
Editor T. Nguyen.
The authors are with the School of Electronic and Information Engi-
neering, Xi’an Jiaotong University, Xi’an, Shaanxi 710049, China (e-mail:
liugz@xjtu.edu.cn; chenls@mail.xjtu.edu.cn).
Color versions of one or more of the figures in this paper are available
online at http://ieeexplore.ieee.org.
Digital Object Identifier 10.1109/TCSVT.2010.2056970
points (DSCPs) [3], [4]. However, as in [6], there is no
policing algorithm involved at the DS domain edge to check
the packets’ importance, and the importance indication of a
packet may be disregarded if legacy packet markers, such as
two rate three color marker (TRTCM), is used [5]. Then two
markers delivery system are proposed in [6]–[8] with a source
marker at the application layer to pre-mark the video packets
to certain source marks and a special edge router marker to
mark the packets at the DS edge according to the current
network congestion state. However, there exist various video
coders, with different characteristics having great impact on
the importance of their compressed data. And the network is
a shared medium, in which it is very difficult to design one
particular edge router marker for each coder. So, in this letter,
we will propose a general two markers delivery system for
streaming applications, which pre-marks importance indices
of compressed data at coding in the application layer, maps
those importance indices to different source marks according
to a simple rule, and marks those packets to DSCPs using
improved two rate three color maker (ITRTCM) at the DS
domain edge in terms of the current network condition. In
this system, streaming applications run over real-time transport
protocol (RTP) and all those packets having extended RTP
headers record the source marks of the video packets. At last,
the core routers in the DS domains deliver those packets to
the egress routers on the paths of video packets according
to per-hop forwarding behaviors corresponding to their drop
precedence. After several hops of DS domains, delivery of
those video packets at last arrive at their destination.
The remainder of this letter is organized as follows. The
delivery system is presented in Section II, which involves two
examples of pre-marking scheme, the uniform source-marking
scheme, and ITRTCM. The simulation and its experimental
results are presented and discussed in Section III. A brief
conclusion is given in Section IV.
II. General Two Markers Delivery System
A. General Source Marking Scheme
The general source marking scheme consists of two steps.
First, when coding a video sequence (VS) with a special coder,
we should measure the importance of a block of data to be
compressed and pre-mark it with one of the K importance
grades in accordance with the coder used. Second, we should
map the K importance grades to L marks and L will be used
1051-8215/$26.00
c
2010 IEEE
下载后可阅读完整内容，剩余4页未读，立即下载
2020-07-01 上传
2021-02-09 上传
2021-02-20 上传
2020-08-26 上传
2021-02-10 上传
2021-02-22 上传
2020-11-18 上传
2021-03-11 上传
2020-03-29 上传
weixin_38655011
- 粉丝: 9
- 资源: 916
我的内容管理
展开
最新资源
- Unity UGUI性能优化实战：UGUI_BatchDemo示例
- Java实现小游戏飞翔的小鸟教程分享
- Ant Design 4.16.8：企业级React组件库的最新更新
- Windows下MongoDB的安装教程与步骤
- 婚庆公司响应式网站模板源码下载
- 高端旅行推荐：官网模板及移动响应式网页设计
- Java基础教程：类与接口的实现与应用
- 高级版照片排版软件功能介绍与操作指南
- 精品黑色插画设计师作品展示网页模板
- 蓝色互联网科技企业Bootstrap网站模板下载
- MQTTFX 1.7.1版：Windows平台最强Mqtt客户端体验
- 黑色摄影主题响应式网站模板设计案例
- 扁平化风格商业旅游网站模板设计
- 绿色留学H5模板：科研教育机构官网解决方案
- Linux环境下EMQX安装全流程指导
- 可爱卡通儿童APP官网模板_复古绿色动画设计
