TR2022-070

Inverse Regular Perturbation with ML-Assisted Phasor Correction for Fiber Nonlinearity Compensation


    •  Dzieciol, H., Koike-Akino, T., Wang, Y., Parsons, K., "Inverse Regular Perturbation with ML-Assisted Phasor Correction for Fiber Nonlinearity Compensation", Optics Letters, DOI: 10.1364/​OL.460929, Vol. 47, No. 14, pp. 3471-3474, June 2022.
      BibTeX TR2022-070 PDF
      • @article{Dzieciol2022jun,
      • author = {Dzieciol, Hubert and Koike-Akino, Toshiaki and Wang, Ye and Parsons, Kieran},
      • title = {Inverse Regular Perturbation with ML-Assisted Phasor Correction for Fiber Nonlinearity Compensation},
      • journal = {Optics Letters},
      • year = 2022,
      • volume = 47,
      • number = 14,
      • pages = {3471--3474},
      • month = jun,
      • doi = {10.1364/OL.460929},
      • url = {https://www.merl.com/publications/TR2022-070}
      • }
  • MERL Contacts:
  • Research Areas:

    Communications, Machine Learning

Abstract:

We improve an inverse regular perturbation (RP) model using a machine learning technique. The proposed learned RP (LRP) model jointly optimizes step-size, gain and phase rotation for individual RP branches. We demonstrate that the proposed LRP can outperform the corresponding learned digital back-propagation (DBP) method based on a split-step Fourier method (SSFM), with up to 0.75 dB gain in a 800km standard single mode fiber link. Our LRP also allows a fractional step-perspan (SPS) modelling to reduce complexity while maintaining superior performance over a 1-SPS SSFM-DBP.