Modelling the spectro-temporal evolution of a frequency comb in a nonlinear optical fibre with a feedforward neural network
Abstract
We deploy a feedforward neural network to predict the spectro-temporal evolution of
a periodic waveform upon nonlinear fibre propagation and demonstrate efficient probing of the full input-parameter space for on-demand comb generation or significant spectral/temporal focusing occurrence. The computational speed is increased by more than 4 orders of magnitude.
Domains
Optics / PhotonicOrigin | Files produced by the author(s) |
---|