Viterbi algorithm for iterative decoding of parallel concatenated convolutional codes

A low complexity Soft-Input Soft-Output (SISO) module, based on Viterbi algorithm, is proposed. This module is suitable for turbo decoding of Parallel Concatenated Convolutional Codes (PCCCs). The interest of this approach resides on the fact that the Viterbi algorithm is used unmodified, still accepting as its inputs soft information. The hard output of the Viterbi algorithm is used to compute the extrinsic information based on the knowledge of systematic and a priori information. Also, a set of scaling factors is used to improve the quality of the extrinsic information. The scaling factors are obtained based on a semi-analytical method. The convergence properties and performance of the proposed turbo decoder are evaluated using EXtrinsic Information Transfer (EXIT) charts and Bit Error Rate (BER) simulations.