Picture for Jamal Darweesh

Jamal Darweesh

Telecom-Paris

Few-bit Quantization of Neural Networks for Nonlinearity Mitigation in a Fiber Transmission Experiment

Add code
May 25, 2022
Jamal Darweesh, Nelson Costa, Antonio Napoli, Bernhard Spinnler, Yves Jaouen, Mansoor Yousefi, .
Figure 1 for Few-bit Quantization of Neural Networks for Nonlinearity Mitigation in a Fiber Transmission Experiment
Figure 2 for Few-bit Quantization of Neural Networks for Nonlinearity Mitigation in a Fiber Transmission Experiment
Figure 3 for Few-bit Quantization of Neural Networks for Nonlinearity Mitigation in a Fiber Transmission Experiment

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.

* 4 pages ,3 figuers 
Viaarxiv icon
Github IconAccess Paper or Ask Questions
Twitter IconFacebook IconLinkedin IconWhatsapp IconMessenger Icon