TITRE Adaptive Packet Video Streaming Over IP Networks - LaBRI

Chapter 4
4 Adaptive Packet Video Transport over IP
Networks
Rapid Advance in digital video coding and networking technologies is leading to the
development of a wide range of new generation audiovisual services and technologies to support
them. Such applications include wired / wireless videoconferencing, interactive digital TV, remote
diagnosis / surgery, distance / remote sensing, process monitoring and tele-education. Many of
these applications involve the use of enabling media coding and content analysis techniques for
media compression, indexing, searching and retrieval.
One of the major requirement for the implementation of such applications is the efficient
transmission of multimedia content (audio, video, text, image) over a broad rang of
telecommunications infrastructure, in particular IP networks such as the Internet.
Multimedia applications have a strict bandwidth, delay, jitter, and loss requirements, which the
current IP networks does not guarantee. Today, IP Quality of Service (QoS) mechanisms and
architecture (Diffserv, Intserv) are expected to address these issues and enable a wide spread of use
of real time IP services. Unfortunately, these QoS control models are not sufficient since they
perform on per-IP domain, and not on end-to-end basis. Promising Service Level Agreement (SLA)
should address this service provisioning, but in the context of mobile IP multimedia service, the
SLA is hard to cope by since there may not be enough resources available in some part of the
network as the terminal is moving into.
Therefore, it is important that the multimedia applications be adaptive to system and network
resource constraints while insuring that end-user requirements are met.
The key contribution of this chapter is lying in the combination of media content analysis
techniques and network control mechanisms for adaptive video streaming transport over IP
networks. In order to meet this goal, we have identified the following sub-goals that are addressed
and evaluated.
• We design a content-based video classification model for translation from video
application level QoS (e.g. MPEG-4 Object Descriptor and / or MPEG-7
Framework) to network system level QoS (e.g. IP Diffserv) (see Section 4.1).
• We design a robust and adaptive application level framing protocol with video stream
multiplexing and unequal error protection (see Section 4.2).
63