Statistical analysis-based approach for check node processor inputs of NB-LDPC decoder

Non-Binary LDPC codes are nowadays considered as a potential competitor of both binary LDPC and convolutional Turbo Codes, mainly when codes with short and moderate codeword lengths are used. However, the Non-Binary nature of these codes makes their decoding process suffers from high computational complexity. This paper deals with the reduction of this computational complexity at the Check Node level based on the selection of the most reliable messages effectively impacting the CN processing. Two main message truncation schemes are proposed leading to a reduction of the memory size needed to store the extrinsic reliability messages of the order of 12 %. The Monte Carlo simulation results show that these truncation schemes do not degrade the performance of the code where in the worst case a loss of 0.06 dB is obtained.