Presentation ideas and initial resources

(Very loosely classified)

Communication problems:
1. Slepian-Wolf theorem: Distributed source coding (Primary resource: Cover&Thomas)
2. Multiple access channel: Multiple people sharing the same communication channel (Primary resource: Cover&Thomas)
3. (Degraded) broadcast channel: One transmitter broadcasting to multiple receivers (Primary resource: Cover&Thomas)
4. Wyner-Ziv: Lossy distributed source coding with side information at the receiver (Primary resource: Cover&Thomas)
5. Writing on dirty paper: Communicating over a channel when transmitter knows the state, but not the receiver (digital watermarking) (Paper)

Models:
1. Arbitrarily-varying channels: Reliable communication under channel uncertainty (Paper)
2. Online channels: Codes against online adversaries (Paper)
3. Network coding channels: Codes against network jamming (Paper)
4. Insertion-deletion channels: Channels with synchronization errors (Paper)
5. Codes for asymmetric errors: Codes for flash memories (Paper)
6. Zero-error channel codes: Channel codes for zero-error channels -- very graph-theoretic (Paper)
7. Communication complexity: Communication for computation of functions (Book)
8. Quantum channels: Communication over quantum channels (Book)

9. Wiretap channels: Information-theoretic secrecy (Paper)

Codes/Bounds:
1. Polar codes: Efficient channel decoding via successive cancellation (Paper)
2. LP decoding of Expander codes: Efficient decoding of channel codes via solving LPs (Paper)
3. List-Decoding RS codes beyond the minimum distance: "Improving" on the Singleton bound (Paper)
4. LP bound for binary codes: Upper bound for binary worst-case channel (Class notes)

5. Bound on Huffman code performance: Improved upper bound for Huffman codes (Paper)

Other problems:
1. Group testing: Identification of a small subset of defectives from a large pool via careful tests (Book)
2. Kolmogorov complexity and information theory: Computational complexity of strings, and connection with entropy (Cover&Thomas)
3. Universal portfolios: How to make money by mixing expert opinions (Cover&Thomas)
4. Gambling and information theory: How to use side-information to make money (Cover&Thomas)
5. Compressive sensing (orthogonal matching pursuit): How to reconstruct a sparse vector via just dot-products (Paper) 
6. Set of entropic vectors: Characterizing the set of all possible types of random variables (Yeung)

Date:  12:30:00 - 14:30:00 - Tuesday, 06, December 2011

Room ERB1009