Quantifying the Operational Benefits of Deep Learning based Dynamic Traffic Prediction using Real-World Dataset

D. Uzunidis; C. Christofodis; I. De Francesca; J. M.Rivas Moscoso; D. Larrabeiti; J. M. Fabrega; D. Marom; I. Tomkos

doi:10.1364/ofc.2025.w2a.46

Quantifying the Operational Benefits of Deep Learning based Dynamic Traffic Prediction using Real-World Dataset

D. Uzunidis^*, C. Christofodis, I. De Francesca, J. M.Rivas Moscoso, D. Larrabeiti, J. M. Fabrega, D. Marom, I. Tomkos

^*Corresponding author for this work

The Institute of Applied Physics

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

A convolutional neural network is trained using real-world data, for dynamic prediction of the required transceivers supporting 6G Χ-haul, leading to 20% and 16% lower average transceiver utilization over static and semi-static cases, respectively.

Original language	English
Title of host publication	Optical Fiber Communication Conference in Proceedings Optical Fiber Communication Conference, OFC 2025 and Optical Fiber Communication Conference (OFC) Postdeadline Papers 2025
Publisher	Optical Society of America
ISBN (Electronic)	9781557527370
DOIs	https://doi.org/10.1364/ofc.2025.w2a.46
State	Published - 2025
Event	2025 Optical Fiber Communication Conference, OFC 2025 - San Francisco, United States Duration: 30 Mar 2025 → 3 Apr 2025

Publication series

Name	Optical Fiber Communication Conference in Proceedings Optical Fiber Communication Conference, OFC 2025 and Optical Fiber Communication Conference (OFC) Postdeadline Papers 2025

Conference

Conference	2025 Optical Fiber Communication Conference, OFC 2025
Country/Territory	United States
City	San Francisco
Period	30/03/25 → 3/04/25

Bibliographical note

Publisher Copyright:
© Optica Publishing Group 2025.

Access to Document

10.1364/ofc.2025.w2a.46

Cite this

Uzunidis, D., Christofodis, C., De Francesca, I., Moscoso, J. M. R., Larrabeiti, D., Fabrega, J. M., Marom, D., & Tomkos, I. (2025). Quantifying the Operational Benefits of Deep Learning based Dynamic Traffic Prediction using Real-World Dataset. In Optical Fiber Communication Conference in Proceedings Optical Fiber Communication Conference, OFC 2025 and Optical Fiber Communication Conference (OFC) Postdeadline Papers 2025 (Optical Fiber Communication Conference in Proceedings Optical Fiber Communication Conference, OFC 2025 and Optical Fiber Communication Conference (OFC) Postdeadline Papers 2025). Optical Society of America. https://doi.org/10.1364/ofc.2025.w2a.46

Uzunidis, D. ; Christofodis, C. ; De Francesca, I. et al. / Quantifying the Operational Benefits of Deep Learning based Dynamic Traffic Prediction using Real-World Dataset. Optical Fiber Communication Conference in Proceedings Optical Fiber Communication Conference, OFC 2025 and Optical Fiber Communication Conference (OFC) Postdeadline Papers 2025. Optical Society of America, 2025. (Optical Fiber Communication Conference in Proceedings Optical Fiber Communication Conference, OFC 2025 and Optical Fiber Communication Conference (OFC) Postdeadline Papers 2025).

@inproceedings{0700ef721261489490f58b1908ad59df,

title = "Quantifying the Operational Benefits of Deep Learning based Dynamic Traffic Prediction using Real-World Dataset",

abstract = "A convolutional neural network is trained using real-world data, for dynamic prediction of the required transceivers supporting 6G Χ-haul, leading to 20\% and 16\% lower average transceiver utilization over static and semi-static cases, respectively.",

author = "D. Uzunidis and C. Christofodis and \{De Francesca\}, I. and Moscoso, \{J. M.Rivas\} and D. Larrabeiti and Fabrega, \{J. M.\} and D. Marom and I. Tomkos",

note = "Publisher Copyright: {\textcopyright} Optica Publishing Group 2025.; 2025 Optical Fiber Communication Conference, OFC 2025 ; Conference date: 30-03-2025 Through 03-04-2025",

year = "2025",

doi = "10.1364/ofc.2025.w2a.46",

language = "אנגלית",

series = "Optical Fiber Communication Conference in Proceedings Optical Fiber Communication Conference, OFC 2025 and Optical Fiber Communication Conference (OFC) Postdeadline Papers 2025",

publisher = "Optical Society of America",

booktitle = "Optical Fiber Communication Conference in Proceedings Optical Fiber Communication Conference, OFC 2025 and Optical Fiber Communication Conference (OFC) Postdeadline Papers 2025",

}

Uzunidis, D, Christofodis, C, De Francesca, I, Moscoso, JMR, Larrabeiti, D, Fabrega, JM, Marom, D & Tomkos, I 2025, Quantifying the Operational Benefits of Deep Learning based Dynamic Traffic Prediction using Real-World Dataset. in Optical Fiber Communication Conference in Proceedings Optical Fiber Communication Conference, OFC 2025 and Optical Fiber Communication Conference (OFC) Postdeadline Papers 2025. Optical Fiber Communication Conference in Proceedings Optical Fiber Communication Conference, OFC 2025 and Optical Fiber Communication Conference (OFC) Postdeadline Papers 2025, Optical Society of America, 2025 Optical Fiber Communication Conference, OFC 2025, San Francisco, United States, 30/03/25. https://doi.org/10.1364/ofc.2025.w2a.46

Quantifying the Operational Benefits of Deep Learning based Dynamic Traffic Prediction using Real-World Dataset. / Uzunidis, D.; Christofodis, C.; De Francesca, I. et al.
Optical Fiber Communication Conference in Proceedings Optical Fiber Communication Conference, OFC 2025 and Optical Fiber Communication Conference (OFC) Postdeadline Papers 2025. Optical Society of America, 2025. (Optical Fiber Communication Conference in Proceedings Optical Fiber Communication Conference, OFC 2025 and Optical Fiber Communication Conference (OFC) Postdeadline Papers 2025).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Quantifying the Operational Benefits of Deep Learning based Dynamic Traffic Prediction using Real-World Dataset

AU - Uzunidis, D.

AU - Christofodis, C.

AU - De Francesca, I.

AU - Moscoso, J. M.Rivas

AU - Larrabeiti, D.

AU - Fabrega, J. M.

AU - Marom, D.

AU - Tomkos, I.

PY - 2025

Y1 - 2025

N2 - A convolutional neural network is trained using real-world data, for dynamic prediction of the required transceivers supporting 6G Χ-haul, leading to 20% and 16% lower average transceiver utilization over static and semi-static cases, respectively.

AB - A convolutional neural network is trained using real-world data, for dynamic prediction of the required transceivers supporting 6G Χ-haul, leading to 20% and 16% lower average transceiver utilization over static and semi-static cases, respectively.

UR - http://www.scopus.com/inward/record.url?scp=105009228200&partnerID=8YFLogxK

U2 - 10.1364/ofc.2025.w2a.46

DO - 10.1364/ofc.2025.w2a.46

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:105009228200

T3 - Optical Fiber Communication Conference in Proceedings Optical Fiber Communication Conference, OFC 2025 and Optical Fiber Communication Conference (OFC) Postdeadline Papers 2025

BT - Optical Fiber Communication Conference in Proceedings Optical Fiber Communication Conference, OFC 2025 and Optical Fiber Communication Conference (OFC) Postdeadline Papers 2025

PB - Optical Society of America

T2 - 2025 Optical Fiber Communication Conference, OFC 2025

Y2 - 30 March 2025 through 3 April 2025

ER -

Uzunidis D, Christofodis C, De Francesca I, Moscoso JMR, Larrabeiti D, Fabrega JM et al. Quantifying the Operational Benefits of Deep Learning based Dynamic Traffic Prediction using Real-World Dataset. In Optical Fiber Communication Conference in Proceedings Optical Fiber Communication Conference, OFC 2025 and Optical Fiber Communication Conference (OFC) Postdeadline Papers 2025. Optical Society of America. 2025. (Optical Fiber Communication Conference in Proceedings Optical Fiber Communication Conference, OFC 2025 and Optical Fiber Communication Conference (OFC) Postdeadline Papers 2025). doi: 10.1364/ofc.2025.w2a.46

Quantifying the Operational Benefits of Deep Learning based Dynamic Traffic Prediction using Real-World Dataset

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this