Accurate Closed-Form Real-Time EGN Model Formula Leveraging Machine-Learning Over 8500 Thoroughly Randomized Full C-Band Systems

Abstract

We derived an approximate non-linear interference (NLI) closed-form model (CFM), capable of handling a very broad range of optical WDM system scenarios. We tested the CFM over 8500 randomized C-band WDM systems, of which 6250 were fully-loaded and 2250 were partially loaded. The systems had highly diversified channel formats, symbol rates, fibers, as well as other parameters. We improved the CFM accuracy by augmenting the formula with simple machine-learning factors, optimized by leveraging the system test-set. We further improved the CFM by adding a term which models special situations where NLI has high self-coherence. In the end, we obtained a very good match with the results found using the numerically-integrated Enhanced GN-model (or EGN-model). We also checked the CFM accuracy by comparing its predictions with full-C-Band splitstep simulations of 300 randomized systems. The combined high accuracy and very fast computation time (milliseconds) of the CFM potentially make it an effective tool for real-time physicallayer-aware optical network management and control.

A QoT Estimation Method using EGN-assisted Machine Learning for Network Planning Applications

“…In Fig. 2, the cumulative distribution function (CDF) of the SNR error in dB of the ML model is compared to that of the closed-form EGN approximate model [13] , henceforth referred to as GN. A mean error of less than 0.1 dB SNR is obtained for the ML-based QoT estimator and 0.15 dB for the GN.…”

Roadmap on optical communications

“…The nonlinear distortion in the few-mode fiber has been studied in [175]. The stimulated Raman scattering is considered to improve the accuracy of the EGN model [176]. To practically and accurately estimate the nonlinear system performance from the device nonlinear characteristics, a new test signal other than the conventional single-tone or multi-tone signal is expected.…”

Multi-wavelength transponders for high-capacity optical networks: a physical-layer-aware network planning study

“…The configurations are chosen in order to minimize the number of required lightpaths (LPs), with candidate configurations being chosen based on the linear SNR. In a second step, nonlinear interference is computed using the closed-form GN model [25]. The SNR takes into account OSNR TX , ASE noise from the amplifiers, and interference due to the fiber nonlinearities.…”