Abstract:
LDPC codes are distinguished by their very efficient decoding algorithms, hence they have a high potential to replace the currently used codes. But this good feature comes with a price: the existence of pseudo-codewords.
In this survey talk, I shall discuss the basic ideas behind the decoding algorithms, the problem, and different ways to characterize the pseudo-codewords. This talk is based on a joint work with R. Koetter, P. Vontobel, and J. Walker in the August issue of Advances in Math.