| Non-binary Low Density Parity Check(NB-LDPC)codes have excellent coding gain and good burst error resistance performance.The Extended Min-Sum(EMS)algorithm is a simplified algorithm proposed to reduce its decoding complexity.Spatially Coupled Low Density Parity Check(SC-LDPC)codes introduce convolutional coupling,further improving coding gain.Windowed Decoding(WD)scheme is the mainstream decoding scheme for non-binary SC-LDPC codes.This paper focuses on three aspects:improving the error correction performance of EMS algorithms,reducing the decoding complexity of traditional WD scheme,and improving the error correction performance of the WD scheme.The main contents are as follows.1.Aiming at the performance loss caused by the truncation of the LogLikelihood Ratio(LLR)of the EMS algorithm,this paper proposes an EMS modification algorithm based on a posteriori probability threshold.By setting a threshold value for the posteriori probability,the variable nodes that need to be corrected most are selected,and correction factors or offsets are introduced more specifically.The simulation results show that at a Bit Error Rate(BER)level of 10-3,the modification EMS algorithm based on a posterior probability threshold can achieve a performance improvement of 0.08dB compared to the traditional modification EMS algorithm;And analysis shows that this performance improvement does not come at the expense of complexity.2.Aiming at the problem of high decoding complexity in traditional WD scheme,this paper proposes a WD based on the EMS algorithm(EMSWD)scheme,which combines the EMS algorithm.Analysis shows that the EMS-WD scheme can reduce the computational complexity by more than 30%compared to traditional WD scheme.In addition,in view of the performance loss of the EMS-WD scheme,this paper proposes an improved EMS-WD(iEMS-WD)scheme by introducing a correction idea based on a posterior probability threshold.Simulation results show that the improved scheme can reduce the performance gap between the EMS-WD scheme and the traditional WD scheme from 0.6dB to 0.3dB at a BER level of 10-4 with almost no complexity cost.3.Aiming at the performance loss of the WD scheme due to the use of sliding window decoders,a scaling windowed decoding(Scaling-WD)scheme with improved error correction performance is proposed,which reduces the LLRs of the unreliable window passed to the next window to achieve improved error correction performance of the WD scheme.The simulation results show that the Scaling-WD scheme can achieve a performance improvement of 0.08dB at a BER level of 10-5,while the complexity cost of this improvement is small.In addition,this paper also applies the above approach to the iEMS-WD scheme,further improves the performance of the iEMS-WD scheme by 0.01 dB when the BER level is 10-4. |