Blind identification of channel codes based on neural network in optical fiber communication
We propose three neural network architectures to recognize Hamming codes and BCH codes with different parameters in the presence of codeword asynchrony at the receiver end in optical fiber communication. The identification performance of these algorithms is compared. Considering practical communication scenarios, random sequences are introduced as interference. Simulation results show that the proposed neural network models can identify the channel coding types under DP-16QAM optical fiber communication transmission.
Authors
- Yiqun Pan (ORCID: https://orcid.org/0000-0003-4890-4378)
- Leijing Yang (ORCID: https://orcid.org/0000-0001-6505-8260)
- Xiangjun Xin (ORCID: https://orcid.org/0000-0002-1690-056X)
- Ming Qin (ORCID: https://orcid.org/0000-0002-8424-6299)
- Sitong Zhou
- Qinghua Tian (ORCID: https://orcid.org/0000-0002-1169-9078)
- Ziqi Liu (ORCID: https://orcid.org/0000-0003-2067-5710)
- Feng Tian
- Yongjun Wang (ORCID: https://orcid.org/0000-0002-9112-816X)
Institutions
- Beijing Institute of Technology (CN)
- Beijing University of Posts and Telecommunications (CN)
Publication Details
- Published
- 2025-03-17
- DOI
- https://doi.org/10.1117/12.3060588
- Primary Topic
- Blind Source Separation Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.00