Deep Reinforcement Learning for Network Provisioning in Elastic Optical Networks

Ziazet, Junior Momo; Jaumard, Brigitte

doi:10.1109/icc45855.2022.9839228

Search citation statements

Order By: Relevance

Paper Sections

Select...

Background and Methodology1

Citation Types

Supporting

Mentioning

Contrasting

Year Published

2024

2025

Publication Types

Select...

Preprint1

Article1

Relationship

Self Cite0

Independent2

Authors

Journals

Cited by 2 publications

(1 citation statement)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Over time, the agent learns to take the best actions in each situation that maximize its rewards (or minimize penalties) [10]. RL is effective for a variety of tasks in optical networks, including resource allocation (wavelengths and bandwidth), traffic engineering of flows to minimize congestion and resiliency against fault management [11][12][13].…”

Section: Background and Methodologymentioning

confidence: 99%

Reinforcement-Learning based routing for packet-optical networks with hybrid telemetry

Navarro,

Koneva,

Sánchez-Macián

et al. 2024

Preprint

View full text Add to dashboard Cite

This article provides a methodology and open-source implementation of Reinforcement Learning algorithms for finding optimal routes in a packet-optical network scenario. The algorithm uses measurements provided by the physical layer (pre-FEC bit error rate and propagation delay) and the link layer (link load) to configure a set of latency-based rewards and penalties based on such measurements. Then, the algorithm executes Q-learning based on this set of rewards for finding the optimal routing strategies. It is further shown that the algorithm dynamically adapts to changing network conditions by re-calculating optimal policies upon either link load changes or link degradation as measured by pre-FEC BER. Mathematics Subject Classification (2020) MSC code1 · MSC code2 · more

show abstract