The objective of this project is to enable low bitrate transmission of highly compressed video streams over wireless channels which are subject to fading. One of the most important challenges in wireless video transmission is that terrestrial wireless channels are typically bandlimited so that significant compression is required; however, this compression results in low quality reconstructed video as well as bitstreams which are extremely sensitive to channel errors. Traditionally, in order to protect against channel errors, powerful forward error control codes were carefully selected and employed at the expense of a portion of the limited bandwidth. This research addresses the problem of bandwidth expenditure on channel error protection by exploiting the redundancy in a source-coded video stream to perform iterative, soft source decoding. With these improvements in source decoding less of the overall bandwidth must be expended on channel coding. This recovered bandwidth can then be devoted to reducing the compression rate, thereby increasing the quality of the transmitted video.

The iterative video decoder consists of two stages. The first stage is a traditional video decoder with error detection capabilities which employs hard channel values, while the second stage is an adaptive Viterbi-like source decoder employing soft channel values or reliability information from a channel decoder. The project focuses on developing efficient and reliable error detection techniques and on determining appropriate decision metrics for the Viterbi-like decoder which incorporate the soft channel values with knowledge of errors from the error detection process.

The objectives of this project are closely related to the teaching objectives. One primary goal of the teaching plan is to integrate this research in wireless communication into undergraduate courses in order to encourage interest in research and graduate study. The second goal is to develop a hands-on technology camp for middle-school girls to actively promote the study of engineering and, in particular, telecommunications. In order to demystify the field of engineering, projects will involve illustrating the relationship between engineering and everyday life, exposing the students to the applicability and prevalence of engineering.

Agency
National Science Foundation (NSF)
Institute
Division of Computer and Communication Foundations (CCF)
Application #
9875582
Program Officer
Julia Abrahams
Project Start
Project End
Budget Start
1999-08-01
Budget End
2003-05-31
Support Year
Fiscal Year
1998
Total Cost
$208,000
Indirect Cost
Name
Drexel University
Department
Type
DUNS #
City
Philadelphia
State
PA
Country
United States
Zip Code
19104