This thesis deals with problems in the general areas of signal processing
and networking, that come up in the context of building a system to transmit
in real time high-quality video signals over certain high speed
segments of the public Internet. By high-quality we mean signals encoded
at bit rates resulting in the reproduction quality attained by current
TV-broadcasts, i.e., transmission of CCIR601 color signals at bit rates in
the range 4-9 Mbits/sec.
Our presentation is divided into three main chapters:
- One chapter is dedicated to the problem of defining stochastic
models for image subband data. We present a Markov Random Field
(MRF) model for this data as well as an algorithm for, given
a set of observations, identifying parameters of an MRF likely to
have generated the observations. Image subbands have been
empirically found not to behave like iid fields: our model captures
this property by decomposing the field into the product of two
fields, one markovian and stationary, the other independent and
non-stationary.
- Another chapter is dedicated to the problem error-resilient
coding of images. We present data compression algorithms whose
most salient feature is that their performance degrades gracefully
in the presence of erasures: the quality of the decoded images
depends only on the amount of data available for decoding, but not
on which specific portions of the encoded bit stream are available.
These coders represent a significant improvement over the
state-of-the-art in the field.
- A third chapter is dedicated to network modeling and control
problems. Our most important contribution is the definition of a
network/coder interface for IP networks which gathers channel state
information, and then sets parameters of the video coder to maximice
the quality of the signal delivered to the receiver, while remaining
fair to other data or video connections: this interface plays a role
analogous to that of a Leaky Bucket controller, in that it specifies
traffic shape parameters which result in simultaneous good Quality
of Service (QoS) for the source and good network performance. Since
the network is not assumed to provide any form of QoS guarantee,
fundamental to our construction is a Hidden Markov model for the
channel, based on which the interface solves a problem of optimal
stochastic control, to decide how to configure the encoder. Other
contributions are modifications to the standard Internet transport
protocol (TCP), to make it suitable for the transport of
delay-constrained traffic and to gather channel state information,
and the design of an error-resilient video coder. Experimental
studies reveal that the proposed system is able to stream
TV-broadcast quality video signals, among hosts in wide-area networks
connected to the experimental vBNS backbone.