Impact of amplifier noise figure modeling in simulations of impairment-aware all-optical networks

Abstract

In this article, we analyze and quantify the impact of the use of different amplifier noise figure models in impairment-aware optical network simulations. We compare network simulations that use three different models, two of them already known and largely employed in the literature, and a third one that we propose here, which is more appropriate to use for gain-clamped amplifiers. We present simulation results for the network blocking probability for each model, for different amplifier output saturation powers. And we also present simulation results for the distribution of optical powers at the input of each amplifier in the network. From our results we can conclude in which cases one can use a simpler model and in which other cases it is worth using a more elaborate model. We show that if the simplest and most common model is used for any network simulation, the obtained blocking probability is overestimated.