Plasmonically Enhanced Optical Accelerator for Nonlinear Signal Processing Based on Artificial Neural Networks
Tobias Blatter, Amane Zürrer, Yannik Horst, Christos Pappas, George Giamougiannis, Apostolos Tsakyridis, Manuel Kohli, Ueli Koch, Miltiadis Moralis‐Pegios, Nikos Pleros, Juerg Leuthold · 2024
We reconstruct a 48 Gbit/s nonlinearly distorted optical signal using an artificial neural network (ANN). The digital ANN execution exceeds traditional nonlinear equalizers, while its analog acceleration using plasmonic-organic-hybrid modulators surpasses conventional digital linear equalizers.