Parallel concatenated convolutional turbo codes: performance analysis for different interleaving schemes [articol]

Abstract

The Turbo Codes reported to achieve Shannon limits are concatenated recursive systematic codes. The performances of the Turbo Codes are dependent on encoder design, the iterative decoding procedures and the interleaving scheme. There are two classic types of interleaving, block and convolutional, which can be used in Turbo Codes and which operate the data in a fix or a pseudorandom order. In this study, we investigate by simulation an iterative decoding algorithm for Turbo Codes using a classical convolutional interleaver (A) or a block interleavear (B). For the same code parameters, the (B) algorithm assures better results then (A) algorithm, assuming an increased number of corrected errors.