The expression ‘modern coding theory’ refers to a broad family of coding techniques and
decoding algorithms that have been developed over the last twenty years. These techniques allow
to achieve the information-theoretic limits of reliable communication and data storage
in many settings, while keeping computational complexity under control.
Three unifying themes in modern coding theory are the use of sparse-graph constructions,
iterative message-passing decoding algorithms, and probabilistic designs/analysis methods.
All of these ideas have far-reaching applications beyond chanel coding.
This class presents the basic tools for analysis and optimization of iterative coding systems.
Introduces several code ensembles: LDPC, Turbo, RA, Fountain codes.
Discusses optimized ensembles, message passing algorithms, density evolution.
In addition, we will devote several lectures to polar codes. This class of codes
are not based on random graph constructions, but they nevertheless share several
of the above features: approximate decoding, focus on probabilistic ideas.
Class Times and Locations
Tue-Thu 3:00PM - 4:20PM
Building 380, Room 380Y