Abstract
This search describes an iterative decoding of product code using Soft Input Soft Output Maximum Likelihood (SISO ML) decoder. The SISO ML decoder is implementing using Pyndiah's algorithm where the soft output of decoder is an estimation of the log likelihood ratio (LLR) of the binary decision of ML-decoder. The theoretical justification of this algorithm is developed, and the method used for computing the soft output is fully described. Then the implementation of iterative decoder of product codes (turbo product codes) is described.Simulation results (using MATLAB program version 7 ) of turbo product codes are presented. The simulation process includes different cases. These result shows that powerful BER performance could be achieved by using block codes with large code length, and BER performance depends mainly on minimum Hamming distance of the outer code. These results also compare between conventional ML decoder and iterative ML decoder and show the powerful BER performance of iterative decoder at low SNR, that make it used in deep space communication
