A Physics-Guided Learning Framework for Accurate and Generalizable EDFA Gain Spectrum Modeling with FourierKAN
Published in Journal of Lightwave Technology (JLT) — under review [Invited], 2026
A physics-guided FourierKAN framework for accurate and generalizable EDFA gain spectrum modeling. Achieves overall MAE of 0.097 dB (89% error reduction vs. DNN transfer-learning baseline) and strong generalization to unseen operating configurations.
Recommended citation: Q. Qiu, Y. Zhu, S. Xiang, Y. Zhang, J. Wen, Z. Zhang, W. Hu, and Q. Zhuge. (2026). "A Physics-Guided Learning Framework for Accurate and Generalizable EDFA Gain Spectrum Modeling with FourierKAN." Submitted to Journal of Lightwave Technology. (under review, invited).
Download Paper