Course content

Bayesian estimation. Modelling and analysis of components in a generic communication system. Mathematical definitions of information content and channel capacity. Principles for optimal information transfer across various types of channels. Data compression. Principles and methods for practical digital representations. Practical channel coding. Performance assessment relative to information theoretic limits. Belief propagation.

Learning outcome

The purpose of the course is twofold. Firstly, it aims at giving the student an understanding of formal mathematical models of information and communication that enables a quantifiacation of the theoretically optimal performance of a communication system. Secondly, the student should get insigth in how these theoretical limits can be reached through practical algorithms and methods.

Learning methods and activities

Lectures and exercises. Postponed/repeated exams may be oral.

Recommended previous knowledge

TTT4115 Communications or equivalent.

Course materials

David MacKay, Information Theory, Inference, and Learning Algorithms, Cambridge University Press, 2003.

Credit reductions