Multi-Task Learning to Enhance Generalizability of Neural Network Equalizers in Coherent Optical Systems

Sasipim Srivallapanondh; Pedro J. Freire; Ashraful Alam; Nelson Costa; Bernhard Spinnler; Antonio Napoli; Egor Sedov; Sergei K. Turitsyn; Jaroslaw E. Prilepsky

Multi-Task Learning to Enhance Generalizability of Neural Network Equalizers in Coherent Optical Systems

Signal Processing 2023-11-06 v3 Machine Learning

Authors: Sasipim Srivallapanondh , Pedro J. Freire , Ashraful Alam , Nelson Costa , Bernhard Spinnler , Antonio Napoli , Egor Sedov , Sergei K. Turitsyn , Jaroslaw E. Prilepsky

View on arXiv ↗ PDF ↗

Abstract

For the first time, multi-task learning is proposed to improve the flexibility of NN-based equalizers in coherent systems. A "single" NN-based equalizer improves Q-factor by up to 4 dB compared to CDC, without re-training, even with variations in launch power, symbol rate, or transmission distance.

Keywords

multi-task learning neural network equalizer coherent optical system q-factor cdc

Cite

@article{arxiv.2307.05374,
  title  = {Multi-Task Learning to Enhance Generalizability of Neural Network Equalizers in Coherent Optical Systems},
  author = {Sasipim Srivallapanondh and Pedro J. Freire and Ashraful Alam and Nelson Costa and Bernhard Spinnler and Antonio Napoli and Egor Sedov and Sergei K. Turitsyn and Jaroslaw E. Prilepsky},
  journal= {arXiv preprint arXiv:2307.05374},
  year   = {2023}
}

Comments

4 pages, European Conference on Optical Communication (ECOC)

Multi-Task Learning to Enhance Generalizability of Neural Network Equalizers in Coherent Optical Systems

Abstract

Keywords

Cite

Comments

Related papers