Machine Learning for Optical Performance Monitoring from Directly Detected PDM-QAM Signals
Konferenz: ECOC 2016 - 42nd European Conference on Optical Communication
18.09.2016 - 22.09.2016 in Düsseldorf, Deutschland
Tagungsband: ECOC 2016
Seiten: 3Sprache: EnglischTyp: PDF
Persönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt
Autoren:
Wass, Jesper; Thrane, Jakob; Piels, Molly; Diniz, Julio C. M.; Jones, Rasmus; Zibar, Darko (DTU Fotonik, Technical University of Denmark, Build. 343, 2800, Denmark)
Inhalt:
Supervised machine learning methods are applied and demonstrated experimentally for inband OSNR estimation and modulation format classification in optical communication systems. The proposed methods accurately evaluate coherent signals up to 64QAM using only intensity information.