| Acronym: ML-PropWin |
| Main Objective: Accurate modelling of mobile radio propagation channels is a crucial factor in the development of wireless communication networks, particularly in urban and complex environments. This project aims to integrate artificial intelligence (AI) techniques, such as generative AI and symbolic regression, to enhance the modelling of these channels, with a focus on predicting propagation loss and, consequently, analysing mobile network coverage. Symbolic regression can be employed to generate fitting equations based on empirical data, while generative AI can be used to create realistic scenarios and simulate propagation environments. Additionally, in a more recent approach, it can be utilised as a generative model to propose equations, which are then evaluated based on how well they fit measurement data. The proposed approach, therefore, aims to find new propagation models and improve the accuracy of signal level predictions compared to traditional models, such as COST, 3GPP, and ITU-R, especially for current and future wireless communication networks (5G and 6G). |
| Region of Intervention: Brazil |
| Reference: 445178/2024-8 |
| Funding: CNPq |
| EU Funding Discrimination: Not applicable |
| Approval Date: 23-12-2024 |
| Start Date: 01-01-2025 |
| End Date: 31-12-2026 |
| Team: Rafael Ferreira da Silva Caldeirinha, Nuno Ricardo Cordeiro Leonor |
| Groups: Antennas and Propagation – Lr |
| Partners: Universidade Federal de Minas Gerais (UFMG), Brazil, Universidade Federal São João Del-Rei (UFSJ), Brazil, Lappeenranta University of Tecnology (LUT), Finland, Pontifício Universidade Católica do Rio de Janeiro (PUC-Rio), Brazil, University of Oulu (OULU), Finland, Universidade de Brasília (UnB), Brazil, Centro Federal de Educação de Minas Gerais (CEFET/MG), Brazil |
| Local Coordinator: Rafael Ferreira da Silva Caldeirinha |
|
|
|
|
|
Associated Publications
|
