RAMONaaS (Resource Allocation Methods for Optical Networks as a Service) is a joint research team between
Universidad de la República (Uruguay)
Federal University of Rio de Janeiro -- UFRJ (Brazil)
IMT Atlantique (France)
The project was conceived in 2022 and officially started in January 2023, being financially supported for 2 years up until December 2024.
Goal
Provide a framework facilitating slicing on the fly to answer “Beyond 5G” (B5G) services requirements.
To achieve such a goal, the project is divided into three work packages:
(WP1) Model-driven mechanisms: We will focus on analytical methods for solving the problem of on-demand resource allocation with a focus on the optical access network, as stated before, and with stochastic demand. We will focus on approaches using Markov Decision Processes.
(WP2) Data-driven mechanisms: In this part of the project, we will focus on AIML methods, in particular reinforcement learning, Neural Network based methods, and deep reinforcement learning to solve problems formulated in this first part of the project at a large scale and for heterogeneous service requirements.
(WP3) Testbed and Integration: This WP will be in charge of the construction and the consolidation of a data set, the implementation of the testbed, including either the 5G air interface transmission or emulated radio access network functionalities. Finally, in this WP, we will integrate the algorithms developed in WP1 and WP2.
IMT Atlantique
Alexandre Reiffers-Masson (Associate Professor, IMT Atlantique)
Luiz Anet Neto (Associate Professor, IMT Atlantique)
Olivier Serge Tsemogne-Kamguia (Postdoc, IMT Atlantique)
Ana Fernanda QUARESMA BATISTA SANTOS (Engineer, IMT Atlantique)
Universidad de la República
Alberto Castro (Associate Professor, UdelaR)
Claudina Rattaro (Associate Professor, UdelaR)
Lucas Inglés (PhD student, UdelaR)
Federal University of Rio de Janeiro (UFRJ)
Daniel Sadoc Menasché (Associate Professor, UFRJ)
José Ferreira de Rezende (Associate Professor, UFRJ)
Lucas Lopes Felipe (PhD student, UFRJ)
Papers
In publications, we acknowledge the project by stating that
This work was partially financed by:
a STIC/AMSUD project between CAPES/BR (88881.694462/2022-01);
Ministry for Europe and Foreign Affairs/FR;
Campus France/FR and the National Agency for Research;
Innovation/UY (MOV_CO_2022_1_1011515).
In 2024, we have published the following work at IFIP Networking 2024:
Eshwar, S. R., Felipe, L. L., Reiffers-Masson, A., Menasché, D. S., & Thoppe, G. (2024). Online Learning of Weakly Coupled MDP Policies for Load Balancing and Auto Scaling. arXiv preprint arXiv:2406.14141.
Visits from Uruguay to France:
Lucas Inglés visits IMT Atlantique from 20th of Sep. to 20th of Oct. 2023
Claudina Rattaro visits IMT Atlantique from 23th to 30th of Sep. 2023
Visits from Brazil to France:
Jose Rezende visits IMT Atlantique from 24th to 30th of Sep. 2023
Lucas Lopes Felipe visits IMT Atlantique from November 2023 to August 2024
Jose Rezende, Fabio David and Daniel Sadoc Menasché visit IMT Atlantique from 24 to 31 of October 2024
Visits from France to Brazil and Uruguay:
Olivier Serge Tsemogne-Kamguia visits UFRJ and UDELAR from 26/11/2023-12/12/2023
Ana Fernanda QUARESMA BATISTA SANTOS visits UFRJ from November 26/11/2023-15/12/2023