Department of Electrical and Computer Engineering
University of Toronto

ECE1508S: Special Topics in Communications:

Network Information Theory

Spring 2014

---

 

Course Description: This course covers the fundamental limits of communication and the techniques to achieve these limits when multiple transmitters and receivers interact in a shared communication medium. We aim to provide a comprehensive coverage of known results and main techniques in network information theory. We pay particular attention to optimization techniques in the context of multiuser communication systems. Course topics include:

 

• Review of classical channel capacity and rate-distortion theorems and the asymptotic equipartition theorem (AEP).
• Gaussian channels with intersymbol interference. Generalized decision-feedback equalization. Orthogonal frequency-division multiplex. Water-filling.
• Fading channel; Capacity and channel state information; Adaptive modulation. MIMO channel.
• Coded modulation for the Gaussian channel.
• Multiple-access channel. Iterative water-filling.
• Channel with side information. Writing-on-dirty-paper channel. Binning.
• Degraded and non-degraded broadcast channels. Marton region. Uplink-downlink duality. Gaussian vector broadcast channel capacity.
• Interference channel. Han-Kabayashi region. Capacity of two-user Gaussian interference channel within 1/2 bit.
• Relay channel. Cover-El Gamal relay strategies. Capacity of multicast relay network within a constant gap.
• Rate-distortion problem. Slepian-Wolf theorem. Wyner-Ziv theorem. CEO problem.

 

This course is of interests to Ph.D. students in the communications group.

Pre-requisite: A first course in information theory: ECE1502 or equivalent.

Grades:  Exam (50%), Project (50%).

Textbooks:

• Abbas El Gamal and Young-Han Kim: Network Information Theory, Cambridge, 2011.
• Thomas M. Cover and Joy A. Thomas, Elements of Information Theory, 2^nd edition. John Wiley, 2006.

Last Updated: 1/15/14