Using Multi-Layer Perceptron to Estimate the OSNR of Unestablished Lightpaths

We investigate a multi-layer perceptron framework to predict the OSNR of unestablished lightpaths considering interference effects and optical monitoring uncertainties. It achieves more than 99% accuracies when the errors are less than 0.5 dB.

[1]  Yvan Pointurier,et al.  Design of low-margin optical networks , 2017, IEEE/OSA Journal of Optical Communications and Networking.

[2]  Luca Barletta,et al.  Machine-learning method for quality of transmission prediction of unestablished lightpaths , 2018, IEEE/OSA Journal of Optical Communications and Networking.

[3]  P. Poggiolini.  The GN Model of Non-Linear Propagation in Uncompensated Coherent Optical Systems , 2012, Journal of Lightwave Technology.

[4]  K. Christodoulopoulos,et al.  Estimating QoT of unestablished lightpaths , 2016, 2016 Optical Fiber Communications Conference and Exhibition (OFC).

[5]  Emmanouel A. Varvarigos,et al.  On reducing optical monitoring uncertainties and localizing soft failures , 2017, 2017 IEEE International Conference on Communications (ICC).