Tian, Q., Pan, Y., Xin, X., Zhu, L., Li, Z., Wang, C., Dong, Z., Gao, R., Tian, F., Wang, F., Yang, L., Zhang, Q., & Wang, Y. (2025). Graph model-aided optimal iterative decoding technique for LDPC in optical fiber communication. Optics Express, 33(1), 1198-1211. https://doi.org/10.1364/OE.534637
Tian, Qinghua ; Pan, Yiqun ; Xin, Xiangjun et al. / Graph model-aided optimal iterative decoding technique for LDPC in optical fiber communication. In: Optics Express. 2025 ; Vol. 33, No. 1. pp. 1198-1211.
@article{c565540757604db6b6c925aa405197cc,
title = "Graph model-aided optimal iterative decoding technique for LDPC in optical fiber communication",
abstract = "The neural networks offer iteration capability for low-density parity-check (LDPC) decoding with superior performance at transmission. However, to cope with increasing code length and rate, the complexity of the neural network increases significantly. This is due to the large amount of feature extraction required to maintain the error correction capability. Based on this gap, we design a new iterative LDPC decoding technique named graph model neural network-belief propagation (GMNN-BP). GMNN-BP uses graph models as a link between deep learning and belief propagation (BP) algorithms, combining the advantages of both. Compared to traditional fully connected neural network decoders, the GMNN-BP decoding has the substantial benefit of avoiding learning and judging codeword categories directly from a large amount of data and requiring less training data as well. The proposed algorithm is verified by simulation and experiment and is tested by using IEEE 802.3ca standard LDPC code word. The results show that the GMNN-BP decoding algorithm is superior to the BP-based iterative decoding method under the same number of iterations, and the maximum gain can reach 1.9dB. When achieving the same performance, the GMNN-BP decoding algorithm only requires half the number of iterations of other algorithms.",
author = "Qinghua Tian and Yiqun Pan and Xiangjun Xin and Lei Zhu and Zhipei Li and Chenchen Wang and Ze Dong and Ran Gao and Feng Tian and Fu Wang and Leijing Yang and Qi Zhang and Yongjun Wang",
note = "Publisher Copyright: {\textcopyright} 2025 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement.",
year = "2025",
month = jan,
day = "13",
doi = "10.1364/OE.534637",
language = "English",
volume = "33",
pages = "1198--1211",
journal = "Optics Express",
issn = "1094-4087",
publisher = "Optica Publishing Group",
number = "1",
}
Tian, Q, Pan, Y, Xin, X, Zhu, L, Li, Z, Wang, C, Dong, Z, Gao, R, Tian, F, Wang, F, Yang, L, Zhang, Q & Wang, Y 2025, 'Graph model-aided optimal iterative decoding technique for LDPC in optical fiber communication', Optics Express, vol. 33, no. 1, pp. 1198-1211. https://doi.org/10.1364/OE.534637
Graph model-aided optimal iterative decoding technique for LDPC in optical fiber communication. / Tian, Qinghua; Pan, Yiqun
; Xin, Xiangjun et al.
In:
Optics Express, Vol. 33, No. 1, 13.01.2025, p. 1198-1211.
Research output: Contribution to journal › Article › peer-review
TY - JOUR
T1 - Graph model-aided optimal iterative decoding technique for LDPC in optical fiber communication
AU - Tian, Qinghua
AU - Pan, Yiqun
AU - Xin, Xiangjun
AU - Zhu, Lei
AU - Li, Zhipei
AU - Wang, Chenchen
AU - Dong, Ze
AU - Gao, Ran
AU - Tian, Feng
AU - Wang, Fu
AU - Yang, Leijing
AU - Zhang, Qi
AU - Wang, Yongjun
N1 - Publisher Copyright:
© 2025 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement.
PY - 2025/1/13
Y1 - 2025/1/13
N2 - The neural networks offer iteration capability for low-density parity-check (LDPC) decoding with superior performance at transmission. However, to cope with increasing code length and rate, the complexity of the neural network increases significantly. This is due to the large amount of feature extraction required to maintain the error correction capability. Based on this gap, we design a new iterative LDPC decoding technique named graph model neural network-belief propagation (GMNN-BP). GMNN-BP uses graph models as a link between deep learning and belief propagation (BP) algorithms, combining the advantages of both. Compared to traditional fully connected neural network decoders, the GMNN-BP decoding has the substantial benefit of avoiding learning and judging codeword categories directly from a large amount of data and requiring less training data as well. The proposed algorithm is verified by simulation and experiment and is tested by using IEEE 802.3ca standard LDPC code word. The results show that the GMNN-BP decoding algorithm is superior to the BP-based iterative decoding method under the same number of iterations, and the maximum gain can reach 1.9dB. When achieving the same performance, the GMNN-BP decoding algorithm only requires half the number of iterations of other algorithms.
AB - The neural networks offer iteration capability for low-density parity-check (LDPC) decoding with superior performance at transmission. However, to cope with increasing code length and rate, the complexity of the neural network increases significantly. This is due to the large amount of feature extraction required to maintain the error correction capability. Based on this gap, we design a new iterative LDPC decoding technique named graph model neural network-belief propagation (GMNN-BP). GMNN-BP uses graph models as a link between deep learning and belief propagation (BP) algorithms, combining the advantages of both. Compared to traditional fully connected neural network decoders, the GMNN-BP decoding has the substantial benefit of avoiding learning and judging codeword categories directly from a large amount of data and requiring less training data as well. The proposed algorithm is verified by simulation and experiment and is tested by using IEEE 802.3ca standard LDPC code word. The results show that the GMNN-BP decoding algorithm is superior to the BP-based iterative decoding method under the same number of iterations, and the maximum gain can reach 1.9dB. When achieving the same performance, the GMNN-BP decoding algorithm only requires half the number of iterations of other algorithms.
UR - http://www.scopus.com/inward/record.url?scp=85215321841&partnerID=8YFLogxK
U2 - 10.1364/OE.534637
DO - 10.1364/OE.534637
M3 - Article
AN - SCOPUS:85215321841
SN - 1094-4087
VL - 33
SP - 1198
EP - 1211
JO - Optics Express
JF - Optics Express
IS - 1
ER -
Tian Q, Pan Y, Xin X, Zhu L, Li Z, Wang C et al. Graph model-aided optimal iterative decoding technique for LDPC in optical fiber communication. Optics Express. 2025 Jan 13;33(1):1198-1211. doi: 10.1364/OE.534637