EE 520: Topics in Communication: Multi-User Information Theory
Instructors
Prof. Aditya Ramamoorthy (Coover 3222) and Prof. Zhengdao Wang (Coover 3134)
Department of Electrical and Computer Engineering
Email: {adityar, zhengdao} AT iastate DOT edu
Network information theory is concerned with finding the fundamental limits on information transfer over communication networks. In this course we shall consider specific topics in network information theory such as multi-terminal source coding, broadcast channels, multiple-access channels etc. Information theory has by and large been quite successful in providing satisfactory answers to a number of questions over point to point channels. However the field of network information theory has a number of open issues that we will highlight during this course.
This course is aimed at students who have a good grasp of
probability and have taken the course on information theory. Of course this is a
graduate class and the main prerequisite is mathematical maturity. We shall
start with a short refresher on the basics of information theory. A tentative
list of topics is given below.
1. Preliminaries - Properties of entropy, relative entropy and mutual information, typical and jointly-typical sequences and source coding
2. Rate-Distortion
3. Capacity under cost constraints
4. Distributed Source Coding - Slepian-Wolf and Wyner-Ziv problems
5. Coding for channels with state - Gelfand-Pinsker problem
6. Broadcast channels
7. Multiple-access channels
8. Relay and multiple relay channels
9. Issues pertaining to feedback
Course Policies
There shall be no exams in this class. There might be at most one
or two homework assignments. In addition there shall be a project that will
typically be a survey of a particular area. A presentation and a well-written
technical report shall be required. Depending on the number of students
enrolled, we may allow students to do this in groups. Each student who takes
this class is expected to scribe notes in LaTeX. I shall be putting up a
template on the course website.
Your grade shall be primarily based on your presentation and report, the quality
of the scribed notes and class participation.
LaTex template for scribing notes
How to Convert PowerPoint Pictures for Use in LaTeX (useful link)
Announcements
Deadline for submitting one page project write-up is Oct. 16th.
Regular meeting time - TR 2:10-3:30pm in 1016 Coover.
First meeting time - Tues., Aug. 26, 9:30-10:50am, at 1041 Coover. Please come prepared with your schedules. We will decide the regular meeting time for the course, for the remainder of the class.
Course information sheet is available[.pdf]
Lecture Notes
Typical sequences [.pdf]
Jointly typical and conditionally typical sequences, quick overview of basic results in information theory [.pdf]
Rate distortion theory (part 1) [.pdf]
Rate distortion theory (part 2) [.pdf]
Coding for channels with state (part 1) [.pdf]
Coding for channels with state (part 2) [.pdf]
Coding for channels with state (part 3) [.pdf]
Writing on dirty paper [.pdf]
Distributed Source Coding - Slepian-Wolf theorem [.pdf]
Rate distortion with side information - Wyner-Ziv problem [.pdf]
Source coding with side information (part 1) [.pdf]
Broadcast channels (part 1) [.pdf]
Broadcast channels (part 2) [.pdf]
Broadcast channels (part 3) [.pdf]
Overview of probabilistic graphical models [.pdf]