Multi-Span OSNR and GSNR Prediction using Cascaded Learning
We implement a cascaded learning framework leveraging three different EDFA and fiber component models for OSNR and GSNR prediction, achieving MAEs of 0.20 and 0.14 dB over a 5-span network under dynamic channel loading.
Authors
- Giacomo Borraccini (ORCID: https://orcid.org/0000-0002-6920-7873)
- Andrea d’Amico (ORCID: https://orcid.org/0000-0003-4632-2139)
- Tingjun Chen (ORCID: https://orcid.org/0000-0002-5717-5755)
- Daniel C. Kilper (ORCID: https://orcid.org/0000-0003-3542-5335)
- Koji Asahi (ORCID: https://orcid.org/0000-0002-3067-9750)
- Ting Wang (ORCID: https://orcid.org/0000-0002-8354-5254)
- Zehao Wang (ORCID: https://orcid.org/0000-0002-7723-1945)
- Yue-Kai Huang (ORCID: https://orcid.org/0000-0003-2762-9223)
Institutions
- NEC (Japan) (JP)
- Duke University (US)
- Princeton University (US)
- Trinity College Dublin (IE)
Publication Details
- Published
- 2025-01-01
- DOI
- https://doi.org/10.1364/ofc.2025.tu3i.6
- Citations
- 1
- Primary Topic
- Optical Network Technologies
- Type
- article
- Field-Weighted Citation Impact
- 0.73