| title | COST INTERACT - Radio Propagation Modeling in an Urban Scenario using Generative Ray Path Sampling | |||||||
|---|---|---|---|---|---|---|---|---|
| categories |
|
|||||||
| tags |
|
|||||||
| image |
|
|||||||
| permalink | /posts/cost-interact-january-2024/ | |||||||
| redirect_from | /posts/cost-interact-january-2024-presentation/ | |||||||
| description | Presentation slides and code for my talk at the COST INTERACT meeting in Dublin, Ireland. |
The presented work is a collaboration between UCLouvain and UniBo. We investigated the possible use of generative Machine Learning to decrease the computational complexity of Ray Tracing. The aim is to train a model that learns how to generate important (see paper) paths, to avoid the usual exhaustive search through all potential ray paths, that has an exponentially growing computational cost.
{% include slides.html url="/assets/slides/2025-01-27-dublin-presentation.html" %}
If you prefer, PowerPoint and PDF versions are also available.
{% bibliography --cited %}
Available on GitHub:
_slides/2025-01-27-dublin-presentation/main.py{: .filepath}.