Comparison of Neural Network Equalizers for IMDD Optical Transmission
Equalization performance of FNN, RNN, LSTM, BiRNN, BiLSTM and reservoir computing is thoroughly compared in IMDD short-reach optical transmission. Experiments show BiRNN/BiLSTM outperform other four, while reservoir computing require minimum parameters for training.
Authors
- Yiqian Shi
- Xihua Zou (ORCID: https://orcid.org/0000-0002-7240-4229)
- Yixian Dong (ORCID: https://orcid.org/0000-0003-0834-4177)
- Lianshan Yan (ORCID: https://orcid.org/0000-0002-3633-7161)
- Wei Pan (ORCID: https://orcid.org/0000-0001-5529-1827)
Institutions
- Southwest Jiaotong University (CN)
Publication Details
- Published
- 2024-11-02
- DOI
- https://doi.org/10.1109/acp/ipoc63121.2024.10809711
- Citations
- 2
- Primary Topic
- Optical Network Technologies
- Type
- article
- Field-Weighted Citation Impact
- 0.42
Funders
- National Key Research and Development Program of China