Dynamic-scaling photonic reservoir computing via adaptive semiconductor-optical-amplifier nonlinearity
Chinese Optics Letters · 2026 · 24 (8) · 081901
DOI: 10.3788/col202624.081901 ↗Kurz gefasst
Fachartikel in der Zeitschrift Chinese Optics Letters, 2026. Verfasst von Qiao Huang und 7 weiteren Autor:innen. Die Arbeit nennt 25 Literaturangaben.
Abstract (Originalsprache)
We propose a dynamic-scaling photonic reservoir computing (Ds-PRC) architecture that leverages the gain saturation of a semiconductor optical amplifier (SOA) to adaptively control its nonlinear operating regime. By integrating a self-calibrating input scaling mechanism, the system automatically converges to an optimal operating point, eliminating the need for manual tuning. Experimental results demonstrate a classification accuracy of 97.6% on the Modified National Institute of Standards and Technology (MNIST) database at a drive current of only 100 mA, along with robust performance in chaotic time series prediction. The architecture exhibits strong hyperparameter robustness and supports an energy-efficient, low-power mode, offering a practical pathway toward scalable optical neuromorphic hardware.
Zitieren
APA
Huang, Q., Hu, X., Gao, H., Zhao, T., Yang, F., Wu, B., Qiu, K., & Wen, F. (2026). Dynamic-scaling photonic reservoir computing via adaptive semiconductor-optical-amplifier nonlinearity. Chinese Optics Letters, 24(8), 081901. https://doi.org/10.3788/col202624.081901
Vancouver
Huang Q, Hu X, Gao H, Zhao T, Yang F, Wu B, et al. Dynamic-scaling photonic reservoir computing via adaptive semiconductor-optical-amplifier nonlinearity. Chinese Optics Letters. 2026;24(8):081901. doi:10.3788/col202624.081901
BibTeX
@article{Huang2026,
author = {Huang, Qiao and Hu, Xin and Gao, Hanwen and Zhao, Tianfeng and Yang, Feng and Wu, Baojian and Qiu, Kun and Wen, Feng},
title = {Dynamic-scaling photonic reservoir computing via adaptive semiconductor-optical-amplifier nonlinearity},
journal = {Chinese Optics Letters},
year = {2026},
volume = {24},
number = {8},
pages = {081901},
doi = {10.3788/col202624.081901}
}
EUVIMED