Domain Adaptation: the Key Enabler of Neural Network Equalizers in Coherent Optical Systems
Pedro J. Freire, Bernhard Spinnler, Daniel Abode, Jaroslaw E. Prilepsky, Abdallah A. I. Ali, Nelson Costa, Wolfgang Schairer, Antonio Napoli, Andrew D. Ellis, Sergei K. Turitsyn
arXiv:2202.12689v115 citations
Originality Synthesis-oriented
AI Analysis
This addresses the problem of efficient training for neural network equalizers in coherent optical systems, representing an incremental improvement.
The paper tackled the calibration of neural network-based equalizers for real transmissions using synthetic data, achieving up to 99% reduction in training process.
We introduce the domain adaptation and randomization approach for calibrating neural network-based equalizers for real transmissions, using synthetic data. The approach renders up to 99\% training process reduction, which we demonstrate in three experimental setups.