English

Experimental Study of Deep Neural Network Equalizers Performance in Optical Links

Signal Processing 2021-06-25 v1

Abstract

We propose a convolutional-recurrent channel equalizer and experimentally demonstrate 1dB Q-factor improvement both in single-channel and 96 x WDM, DP-16QAM transmission over 450km of TWC fiber. The new equalizer outperforms previous NN-based approaches and a 3-steps-per-span DBP.

Keywords

Cite

@article{arxiv.2106.13133,
  title  = {Experimental Study of Deep Neural Network Equalizers Performance in Optical Links},
  author = {Pedro J. Freire and Yevhenii Osadchuk and Bernhard Spinnler and Wolfgang Schairer and Antonio Napoli and Nelson Costa and Jaroslaw E. Prilepsky and Sergei K. Turitsyn},
  journal= {arXiv preprint arXiv:2106.13133},
  year   = {2021}
}

Comments

Optical Fiber Communication Conference and Exhibition (OFC) 2021 - Oral Presentation

R2 v1 2026-06-24T03:34:00.063Z