Abstract
A neural network is quantized for the mitigation of nonlinear and components’ distortions in a 16-QAM 9x50km dual-polarization fiber transmission experiment. Post-training additive power-of-two quantization at 6 bits incurs a negligible Q-factor penalty. At 5 bits, the model size is reduced by 85%, with 0.8 dB penalty.
© 2022 The Author(s)
PDF ArticleMore Like This
Fan Zhang, Xiansong Fang, and Xinyu Chen
M1H.1 Optical Fiber Communication Conference (OFC) 2022
Prasham Jain, Lutz Lampe, and Jeebak Mitra
We1C.5 European Conference and Exhibition on Optical Communication (ECOC) 2022
Pedro J. Freire, Michael Anderson, Bernhard Spinnler, Thomas Bex, Jaroslaw E. Prilepsky, Tobias A. Eriksson, Nelson Costa, Wolfgang Schairer, Michaela Blott, Antonio Napoli, and Sergei K. Turitsyn
We1C.2 European Conference and Exhibition on Optical Communication (ECOC) 2022