AI-driven Intent-based Optimization with Algorithmic Machine Learning for 6G Networks
Reference number | |
Coordinator | RISE Research Institutes of Sweden AB |
Funding from Vinnova | SEK 98 474 |
Project duration | January 2025 - July 2025 |
Status | Ongoing |
Venture | 6G - Competence supply |
Call | 6G - Supervision of degree work |
Important results from the project
From the research perspective, the project delivered mixed results. We simplified the thesis scope to focus on the MIS problem, relevant for 6G resource allocation. Our system matched state-of-the-art performance of generative AI approaches but could not incorporate natural language constraints into the optimization. While the initial approach did not fully succeed, it laid the groundwork for future research and sparked further collaboration opportunities.
Expected long term effects
The findings will guide future research and have already led to an internal follow-up project. We are establishing formal collaborations with companies and universities to continue exploring this topic, with potential new Vinnova applications. Code is available at https://git.ri.se/daniel.perez/gnn-co and the thesis report at https://lup.lub.lu.se/student-papers/search/publication/9207177.
Approach and implementation
The project followed a data-driven exploratory research methodology. During the first month, the topic of the thesis was more formally defined. The scope was refined to focus on MIS with natural language constraints. The time plan was followed: first phase was the initial literature review and scope definition (End og January to mid March), then the implementation and prototyping phase was performed (mid March – end May), then the results analysis and thesis report writing (end May-July).