Differentiable Machine Learning-Based Modeling for Directly-Modulated Lasers

TL;DR

This paper develops and compares differentiable machine learning surrogate models for directly modulated lasers, enabling end-to-end optical communication system optimization despite the complex, non-analytical laser behavior.

Contribution

It introduces novel ML-based surrogate models for DMLs, addressing the challenge of differentiability in large-signal regimes, and demonstrates their effectiveness in a practical equalization scenario.

Findings

Convolutional attention transformer outperforms other models in accuracy.

Surrogate models achieve low root mean square error.

Models enable end-to-end optimization of optical links.

Abstract

End-to-end learning has become a popular method for joint transmitter and receiver optimization in optical communication systems. Such approach may require a differentiable channel model, thus hindering the optimization of links based on directly modulated lasers (DMLs). This is due to the DML behavior in the large-signal regime, for which no analytical solution is available. In this paper, this problem is addressed by developing and comparing differentiable machine learning-based surrogate models. The models are quantitatively assessed in terms of root mean square error and training/testing time. Once the models are trained, the surrogates are then tested in a numerical equalization setup, resembling a practical end-to-end scenario. Based on the numerical investigation conducted, the convolutional attention transformer is shown to outperform the other models considered.