We consider the problem of forecasting the Quality-of-Transmission (QoT) of deployed lightpaths in a Wavelength Division Multiplexing (WDM) optical network. QoT forecasting plays a determinant role in network management and planning, as it allows network operators to proactively plan maintenance or detect anomalies in a lightpath. To this end, we leverage Bayesian Recurrent Neural Networks for learning uncertainty-aware probabilistic QoT forecasts, i.e., for modelling a probability distribution of the QoT over a time horizon. We evaluate our proposed approach on the open-source Microsoft Wide Area Network (WAN) optical backbone dataset. Our illustrative numerical results show that our approach not only outperforms state-of-the-art models from literature, but also predicts intervals providing near-optimal empirical coverage. As such, we demonstrate that uncertainty-aware probabilistic modelling enables the application of QoT forecasting in risk-sensitive application scenarios.

Di Cicco, N., Talpini, J., Ibrahimi, M., Savi, M., Tornatore, M. (2023). Uncertainty-Aware QoT Forecasting in Optical Networks with Bayesian Recurrent Neural Networks. In Proceedings of IEEE International Conference on Communications 2023 (IEEE ICC 2023) (pp.441-446). IEEE [10.1109/ICC45041.2023.10278767].

Uncertainty-Aware QoT Forecasting in Optical Networks with Bayesian Recurrent Neural Networks

Talpini J.;Savi M.;
2023

Abstract

We consider the problem of forecasting the Quality-of-Transmission (QoT) of deployed lightpaths in a Wavelength Division Multiplexing (WDM) optical network. QoT forecasting plays a determinant role in network management and planning, as it allows network operators to proactively plan maintenance or detect anomalies in a lightpath. To this end, we leverage Bayesian Recurrent Neural Networks for learning uncertainty-aware probabilistic QoT forecasts, i.e., for modelling a probability distribution of the QoT over a time horizon. We evaluate our proposed approach on the open-source Microsoft Wide Area Network (WAN) optical backbone dataset. Our illustrative numerical results show that our approach not only outperforms state-of-the-art models from literature, but also predicts intervals providing near-optimal empirical coverage. As such, we demonstrate that uncertainty-aware probabilistic modelling enables the application of QoT forecasting in risk-sensitive application scenarios.
paper
Forecasting; Machine Learning; Quality-of-Transmission; Regression; Uncertainty;
English
2023 IEEE International Conference on Communications, ICC 2023 - 28 May 2023 through 1 June 2023
2023
Zorzi, M; Tao, M; Saad, W
Proceedings of IEEE International Conference on Communications 2023 (IEEE ICC 2023)
9781538674628
23-ott-2023
2023
2023
441
446
10278767
https://ieeexplore.ieee.org/document/10278767
open
Di Cicco, N., Talpini, J., Ibrahimi, M., Savi, M., Tornatore, M. (2023). Uncertainty-Aware QoT Forecasting in Optical Networks with Bayesian Recurrent Neural Networks. In Proceedings of IEEE International Conference on Communications 2023 (IEEE ICC 2023) (pp.441-446). IEEE [10.1109/ICC45041.2023.10278767].
File in questo prodotto:
File Dimensione Formato  
Di Cicco-2023-ICC-preprint.pdf

accesso aperto

Descrizione: Intervento a convegno
Tipologia di allegato: Submitted Version (Pre-print)
Licenza: Altro
Dimensione 345.75 kB
Formato Adobe PDF
345.75 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/453039
Citazioni
  • Scopus 3
  • ???jsp.display-item.citation.isi??? 1
Social impact