High-Fidelity Prediction of Perturbed Optical Fields using Fourier Feature Networks

TL;DR

This paper introduces a Fourier Feature Network-based machine learning framework that efficiently predicts the effects of physical perturbations on optical channels, achieving high accuracy with fewer parameters and less data.

Contribution

The novel approach encodes perturbations into a Fourier basis, enabling compact models to accurately predict complex optical field changes with minimal data and computational resources.

Findings

Achieved 0.995 complex correlation with ground truth predictions.

Improved accuracy by an order of magnitude over standard networks.

Reduced model complexity by 85% in parameters.

Abstract

Predicting the effects of physical perturbations on optical channels is critical for advanced photonic devices, but existing modelling techniques are often computationally intensive or require exhaustive characterisation. We present a novel data-efficient machine learning framework that learns the perturbation-dependent transmission matrix of a multimode fibre. To overcome the challenge of modelling the resulting highly oscillatory functions, we encode the perturbation into a Fourier Feature basis, enabling a compact multi-layer perceptron to learn the mapping with high fidelity. On experimental data from a compressed fibre, our model predicts the output field with a 0.995 complex correlation to the ground truth, improving accuracy by an order of magnitude over standard networks while using 85\% fewer parameters. This approach provides a general tool for modelling complex optical systems from sparse measurements.