讲座人：Xiong Zixiang，美国德州农工大学教授，IEEE Fellow
Driven by a host of emerging applications, distributed compression has assumed renewed interest in the past decade. Although the Slepian-Wolf theorem has been known for more than 40 years and progresses have been made recently on the rate region of quadratic Gaussian two-terminal source coding, finding the sum-rate bound of quadratic Gaussian multiterminal (MT) source coding with more than two terminals is still an open problem.
In this short course, I'll start with reviewing Shannon's classic rate-distortion theory for single sources before going over existing results on distributed compression problems (e.g., Slepian-Wolf and Wyner-Ziv) and describing a set of new results we obtained recently.
Day 1: Review of R-D theory and Slepian-Wolf coding
Day 2: Wyner-Ziv coding and Multiterminal source coding
1) The generalized Gaussian CEO problem: New cases with tight rate region
2) A new class of quadratic Gaussian MT problems with tight sum-rate
3) The supremum sum-rate loss of quadratic Gaussian MT source coding
4) A new sufficient condition for sum-rate tightness of quadratic Gaussian MT source coding
Day 5: Distributed source coding of linear functions: Partial sum-rate tightness and gap to optimal sum-rate