A Novel Approach for Joint Analytical and ML-assisted GSNR Estimation in Flexible Optical Network
F. Arpanaei, B. Shariati, P. Safari, M. Ranjbar Zefreh, J. A. Hernandez, A. Carena, J. Fischer, D. Larrabeiti
This paper proposed a novel approach to perform QoT estimation relying on joint exploitation of machine learning and analytical formula that offers accurate estimation when applied to scenarios with heterogeneous span profiles and sparsely occupied links.
Dual Time and Frequency Domain Optical Layer Digital Twin
Devigili, M. Ruiz, N. Costa, A. Napoli, J. Pedro, and L. Velasco
This paper demosntrated a digital twin for failure detection in optical networks. Artificial neural networks-based models for optical constellation analysis enable predicting the transmitted signal in the time domain whereas analytical models are normally used to estimate their spectral evolution.
Multi-Agent-based Dynamic Optical Subcarrier Allocation for Near Real-Time P2MP Operation
H. Shakespear-Miles, M. Ruiz, A. Napoli, and L. Velasco
This paper proposed a multi-agent system (MAS) to manage subcarrier allocation in point-to-multipoint connectivity based on Digital Subcarrier Multiplexing. Similar performance to a centralized approach is shown, which allows for near-real time operation with increased scalability