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

2022/05/23 by Jamal Darweesh, Nelson Costa, Darweesh, Jamal +11
Engineering · Physics and Astronomy · #Optical Network Technologies #Advanced Fiber Laser Technologies #Photonic and Optical Devices

paper · pdf · doi:10.48550/arxiv.2205.11284

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.

Related

• Quantization of Neural Network Equalizers in Optical Fiber Transmission Experiments
• Compensation of Fiber Nonlinearities in Digital Coherent Systems Leveraging Long Short-Term Memory Neural Networks
• Nonlinear Interference Mitigation via Deep Neural Networks
• Deep Neural Network Assisted Second-Order Perturbation-Based Nonlinearity Compensation
• QAMNet: Fast and Efficient Optical QAM Neural Networks