Data-Driven Fronthaul Slice Resizing in Next-Generation RAN Networks

Nome do aluno Nilton Flavio Sousa Seixas
Título do trabalho Data-Driven Fronthaul Slice Resizing in Next-Generation RAN Networks
Resumo do trabalho The evolution toward 6G networks introduces unprecedented demands on Radio Access Network (RAN) management, driven by requirements such as near-zero latency, ultra-high reliability, and massive device connectivity. These challenges are further intensified by the dynamic and tidal nature of mobile traffic, which renders static allocation of optical and radio resources impractical and calls for automated, adaptive approaches.
To address this problem, this research proposes a Data-Driven Decision-Making (DDDM) framework for dynamic resource management in 6G RAN networks. The framework integrates Artificial Intelligence (AI), Machine Learning (ML), and big data analytics to enable self-configuration, self-optimization, and self-resizing, in alignment with the Zero-Touch Management (ZTM) paradigm. It is structured into multiple stages, including data definition, extraction, transformation, loading, analysis, decision-making, action, and retraining cycles, ensuring continuous adaptation of resources to fluctuating demands.
Numerical experiments demonstrate the framework’s effectiveness in reducing service blocking, improving optical spectrum utilization, and maintaining synchronization between antennas and Baseband Units (BBUs). Additionally, the study introduces specific algorithms for event-driven traffic scenarios and validates them through simulations in Ultra-Dense Elastic Optical Networks (UD-EONs).
The contributions of this work include the development and validation of the proposed DDDM framework, methodologies for robust dataset preparation and experimental reproducibility, and the identification and solution of key RAN resizing problems. Overall, this research provides a foundation for the deployment of intelligent, fully automated management systems capable of sustaining the scalability and efficiency required by future 6G networks.
Orientador Gustavo Bittencourt Figueiredo
Membro externo 1 (com afiliação) Fabio Luciano Verdi (UFSCar)
Link para o curriculum lattes http://lattes.cnpq.br/9143186843657940
Membro interno 1 (com afiliação) Maycon Leone Maciel Peixoto (UFBA)
Link para o curriculum lattes http://lattes.cnpq.br/5003713680310544
Suplente do membro externo (com afiliação) Helder May N. Oliveira (USP)
Link para o curriculum lattes http://lattes.cnpq.br/1468872219964148
Suplente do membro interno (com afiliação) Cassio V. S. Prazeres
Link para o curriculum lattes http://lattes.cnpq.br/5075736089100544
Data do exame 15 Apr, 2026
Horário do exame 9:00 AM

 

Data da Defesa: 
15/04/2026 - 09:00
Tipo de Defesa: 
Qualificação de Doutorado