Machine learning-based EDFA Gain Model Generalizable to Multiple Physical Devices

09/11/2020

∙

We report a neural-network based erbium-doped fiber amplifier (EDFA) gain model built from experimental measurements. The model shows low gain-prediction error for both the same device used for training (MSE ≤ 0.04 dB^2) and different physical units of the same make (generalization MSE ≤ 0.06 dB^2).

READ FULL TEXT

Machine learning-based EDFA Gain Model Generalizable to Multiple Physical Devices

Path loss, beamforming gain and time dynamics measurements at 28 GHz for 90

Deep Structured Energy-Based Image Inpainting

Experimental validation of machine-learning based spectral-spatial power evolution shaping using Raman amplifiers

Reinforcement Learning Based Speech Enhancement for Robust Speech Recognition

Learned Iterative Decoding for Lossy Image Compression Systems

Low Complexity Convolutional Neural Networks for Equalization in Optical Fiber Transmission

Sibling Neural Estimators: Improving Iterative Image Decoding with Gradient Communication

Machine learning-based EDFA Gain Model Generalizable to Multiple Physical Devices

Related Research

Path loss, beamforming gain and time dynamics measurements at 28 GHz for 90

Deep Structured Energy-Based Image Inpainting

Experimental validation of machine-learning based spectral-spatial power evolution shaping using Raman amplifiers

Reinforcement Learning Based Speech Enhancement for Robust Speech Recognition

Learned Iterative Decoding for Lossy Image Compression Systems

Low Complexity Convolutional Neural Networks for Equalization in Optical Fiber Transmission

Sibling Neural Estimators: Improving Iterative Image Decoding with Gradient Communication