Over-the-fiber Digital Predistortion Using Reinforcement Learning
Paper in proceeding, 2021

We demonstrate, for the first time, experimental over-the-fiber training of transmitter neural networks (NNs) using reinforcement learning. Optical back-to-back training of a novel NN-based digital predistorter outperforms arcsine-based predistortion with up to 60% bit-error-rate reduction.

Author

Jinxiang Song

Chalmers, Electrical Engineering, Communication and Antenna Systems, Communication Systems

Zonglong He

Chalmers, Microtechnology and Nanoscience (MC2), Photonics

Christian Häger

Chalmers, Electrical Engineering, Communication and Antenna Systems, Communication Systems

Magnus Karlsson

Chalmers, Microtechnology and Nanoscience (MC2), Photonics

Alexandre Graell I Amat

Chalmers, Electrical Engineering, Communication and Antenna Systems, Communication Systems

Henk Wymeersch

Chalmers, Electrical Engineering, Communication and Antenna Systems, Communication Systems

Jochen Schröder

Chalmers, Microtechnology and Nanoscience (MC2), Photonics

2021 European Conference on Optical Communications, ECOC 2021

9605972
978-1-6654-3868-1 (ISBN)

47th European Conference on Optical Communications, ECOC 2021
Bordeaux, France,

Unlocking the Full-dimensional Fiber Capacity

Knut and Alice Wallenberg Foundation (KAW 2018.0090), 2019-07-01 -- 2024-06-30.

Subject Categories

Telecommunications

Other Physics Topics

Bioinformatics (Computational Biology)

DOI

10.1109/ECOC52684.2021.9605972

More information

Created

11/30/2021