Domain Adaptation: the Key Enabler of Neural Network Equalizers in Coherent Optical Systems

2022/02/25 by Freire, Pedro J., Spinnler, Bernhard, Abode, Daniel +7
#FOS: Computer and information sciences #FOS: Electrical engineering #Machine Learning (cs.LG) #Signal Processing (eess.SP) #electronic engineering #information engineering

paper · doi:10.48550/arxiv.2202.12689

Abstract

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.

Related

• Multi-Task Learning to Enhance Generalizability of Neural Network Equalizers in Coherent Optical Systems
• Neural Networks-based Equalizers for Coherent Optical Transmission: Caveats and Pitfalls
• Recent Advances on Machine Learning-aided DSP for Short-reach and Long-haul Optical Communications
• Experimental Evaluation of Computational Complexity for Different Neural Network Equalizers in Optical Communications
• Power and Modulation Format Transfer Learning for Neural Network\n Equalizers in Coherent Optical Transmission Systems