Design Optimization of a Double Stage Extended L-band EDFA Using Neural Networks Trained on Experimental Data
Conference: ECOC 2024 - 50th European Conference on Optical Communication
09/22/2024 - 09/26/2024 at Frankfurt, Germany
Proceedings: ITG-Fb. 317: ECOC 2024
Pages: 4Language: englishTyp: PDF
Authors:
Rabbani, Hamed; Jalilpiran, Saber; LaRochelle, Sophie; Rusch, Leslie A.
Abstract:
Designing a double-stage L-band amplifier with low pump powers, low noise figure, and flat gain is challenging. We tackle this with fast and accurate neural networks amplifier models trained on experimental data. Our approaches optimize for flat gain, while minimizing noise figure and pump power.