A simple accurate way to model noise-seeded ultrafast nonlinear processes

Abstract

Noise can play an important role in nonlinear pulse propagation. It is not only the origin of fluctuations in supercontinuum but can also determine the generated signal amplitude and phase, as seen in phenomena such as noise-seeded four-wave mixing and spontaneous Raman scattering. Current models rely on input-pulse shot noise and Raman Langevin term to simulate noise-seeded Kerr effects and spontaneous Raman scattering, respectively. However, they cannot accurately simulate any nonlinear effects exhibiting asymmetries, for example, between spontaneous Stokes and anti-Stokes generation, or in amplification with emission and absorption cross sections. Moreover, combining these two techniques does not lead to accurate modeling of both effects simultaneously, especially in unsaturated Raman operational regimes or supercontinuum where spontaneous Raman emission noise plays a crucial role. To resolve this issue, we propose a new modified shot-noise model by directly introducing noise into the pulse propagation equation. It minimally modifies current models that contain only stimulated terms.