Analysis and design of raptor codes for joint decoding using Information Content evolution

Abstract

This paper is eligible for the student paper award. In this paper, we present an analytical analysis of the convergence of raptor codes under joint decoding over the binary input additive white noise channel (BIAWGNC), and derive an optimization method. We use information content evolution under Gaussian approximation, and focus on a new decoding scheme that proves to be more efficient: the joint decoding of the two code components of the raptor code. In our general model, the classical tandem decoding scheme appears to be a sub-case, and thus, the design of LT codes is also possible.