Queen's Math & Stats Masthead
Math and Eng | Math and Stats | Queen's University

Fady Alajaji, P. Eng.

Professor, Mathematics and Engineering

Department of Mathematics and Statistics
Queen's University at Kingston
Kingston, Ontario, Canada, K7L 3N6
Phone: (613) 533-2423; Fax: (613) 533-2964
E-mail: fa@queensu.ca

Research group: Communications and Information Theory (Applied Mathematics)

picture of me

Research Interests

Keywords : Information theory, joint source-channel coding, digital communications, error-control coding, data compression, applied probability, machine learning.

My primary research interests belong to the general area of information and communication theory. More specifically, I am interested in the coding of information-bearing signals for transmission over noisy communication networks. This problem is addressed at two levels. One objective is to understand and investigate the Shannon theoretic aspect of this problem -- i.e., to determine the fundamental limits of how efficiently one can encode information and still be able to recover it with negligible loss. Another vital objective is to develop efficient coding techniques and algorithms for achieving reliable data transmission over wireless networks. More recently, I have developed an interest in the information bottleneck principle and the use of information-theoretic methods in machine learning.

I also have a long-standing interest in probability problems and their applications, including random processes with reinforcement, contagion models and generalized Polya urns, stochastic modeling and analysis of network epidemics, game theory, and effective bounds for the probability of a union of events under partial information. Details about current research activities follow:

  1. Information Theory: Properties of information measures and their operational characterizations for systems with memory; information spectrum techniques; Shannon coding theorems, source and channel coding; error probability, cutoff rates and reliability function; capacity and capacity-cost function; feedback capacity; data privacy; information hiding; analysis and modeling of channels with memory, burst-noise, finite-state and correlated fading channels, error analysis of communication systems; two-way networks.

  2. Joint Source-Channel Coding: Channel optimized vector quantization; clustering algorithms; bandwidth efficient source-optimized channel codes; applications to wireless communications and sensor networks; source-channel signal mapping and modulation; maximum a posteriori joint source-channel decoding; Turbo codes, low-density parity-check codes and iterative decoding; hybrid digital-analog signaling and source-channel coding; multiple-input multiple-output channels and space-time coding; applications to multimedia processing and communications.

  3. Applied Probability and Machine Learning: Contagion based models, generalized Polya urns, spatio-temporal contagion models for networks and problems in network epidemics; Markov stochastic modeling; game theory and applications; upper and lower bounds on the probability of finite unions of events under limited information. Information-theoretic methods in machine learning; information botleneck; generative-adversarial networks.

Research Publications



Undergraduate Students

Graduate Students

Post-Doctoral Fellows

Event Organizations and Committees

Math. & Eng. Communications and Information Theory Group