E. V. Sedov, I. S. Chekhovskoy, J. E. Prilepsky, “Neural network for calculating direct and inverse nonlinear Fourier transform”, Kvantovaya Elektronika, 2021, Volume 51, Number 12,Pages <nobr>1118

This article is cited in 7 papers

Selection of reports presented at the 8th All-Russian conference on fibre optics (5-8 October 2021, Perm) (compiled and edited by S.L.Semjonov)

Neural network for calculating direct and inverse nonlinear Fourier transform

E. V. Sedov^a, I. S. Chekhovskoy^a, J. E. Prilepsky^b

^a Novosibirsk State University
^b Aston Institute of Photonic Technologies, Aston University, Birmingham, UK

Abstract: A neural network architecture is proposed that allows a continuous nonlinear spectrum of optical signals to be predicted and an inverse nonlinear Fourier transform (NFT) to be performed for signal modulation. The average value of the relative error in predicting the continuous spectrum by the neural network when calculating the direct NFT is found to be 2.68 × 10^-3, and the average value of the relative error in predicting the signal for the inverse NFT is 1.62 × 10^-4.

Keywords: nonlinear Schrödinger equation, inverse scattering problem method, Zakharov–Shabat problem, nonlinear Fourier transform, neural networks, machine learning.

Received: 26.10.2021