Hide menu

Available internal projects

Software for density evolution studies of LDPC codes

Contact:

Description

An LDPC code is a linear binary block code characterized by a parity-check matrix with a low number of nonzero elements i.e. ones. If the number of ones in the columns of the parity check matrix are the same for all columns the LDPC code is said to be regular otherwise it is called irregular. The sparsity of the parity check matrix of an LDPC code makes them ideal objects for iterative probabilistic decoding, with very impressive performance. The preferred way to represent an LDPC code when considering iterative decoding is in the form of a bipartite graph. Using this viewpoint simple yet powerful probabilistic decoding algorithms have been developed for LDPC codes and the study of these codes is one of the hottest topics in coding theory today. The probabilistic behaviour of the iterative decoding algorithm can be studied with the so called density evolution method.

This project aims at studying the density evolution of LDPC codes. The core part of the project will be to implement software (in C++ under the IT++ simulation environment) for performing density evolution for LDPC codes. A firm interest in programming and telecommunications is desirable for the applicant.

References

[1] A. Shokrollahi, LDPC Codes: An Introduction, //www.ics.uci.edu/~welling/teaching/ICS279/LPCD.pdf

Page responsible: Danyo Danev
Last updated: 2015 01 13   15:49