TY - JOUR

T1 - Data-Driven Erbium-doped Fiber Amplifier Gain Modeling Using Gaussian Process Regression

AU - Harvey, Calum

AU - Faruk, Md Saifuddin

AU - Savory, Seb J.

PY - 2024/8/9

Y1 - 2024/8/9

N2 - We propose a data-driven erbium-doped fiber amplifier (EDFA) gain model utilizing Gaussian process regression (GPR). An additive Laplacian and radial-basis function kernel is proposed for the GPR and was found to outperform deep neural network (DNN) methods while additionally providing prediction uncertainty. Performance is measured using mean absolute error (MAE) averaged across five different EDFAs with three manufacturers. The GPR achieves an MAE of 0.1 dB using 30 training samples in contrast to the DNN that achieves an MAE of 0.25 dB using 3000 training samples. Additionally, we demonstrate that active learning can be used to improverobustness and repeatability of convergence.

AB - We propose a data-driven erbium-doped fiber amplifier (EDFA) gain model utilizing Gaussian process regression (GPR). An additive Laplacian and radial-basis function kernel is proposed for the GPR and was found to outperform deep neural network (DNN) methods while additionally providing prediction uncertainty. Performance is measured using mean absolute error (MAE) averaged across five different EDFAs with three manufacturers. The GPR achieves an MAE of 0.1 dB using 30 training samples in contrast to the DNN that achieves an MAE of 0.25 dB using 3000 training samples. Additionally, we demonstrate that active learning can be used to improverobustness and repeatability of convergence.

U2 - 10.1109/LPT.2024.3441110

DO - 10.1109/LPT.2024.3441110

M3 - Article

VL - 36

SP - 1097

EP - 1100

JO - IEEE Photonics Technology Letters

JF - IEEE Photonics Technology Letters

SN - 1041-1135

IS - 18

ER -