Differentiable Ray Tracing with Gaussians for Unified Radio Propagation Simulation and View Synthesis
Niklas Vaara, Lam Huynh, Pekka Sangi, Miguel Bordallo López, Janne Heikkilä · 2026
Read Differentiable Ray Tracing with Gaussians for Unified Radio Propagation Simulation and View Synthesis by Niklas Vaara, Lam Huynh, Pekka Sangi, Miguel Bordallo López, Janne Heikkilä (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-07781.txt.
Description
Niklas Vaara and four co-authors let one neural scene representation serve two jobs that usually live apart: photorealistic view synthesis and radio-frequency propagation simulation. Gaussian splatting scenes render beautifully but are built for appearance, not for the ray geometry that radio digital twins need to compute multi-bounce signal paths and their delays and losses.
Their framework embeds the Gaussian primitives into a hardware-accelerated ray tracing structure, so point-to-point radio paths can be computed between arbitrary locations directly in the visually reconstructed scene. Channel impulse responses come out of visual-only reconstructions. The cross-modal result suggests neural reconstructions can act as a unified spatial representation for both electromagnetic simulation and rendering.
Catalog record
| CREATOR | Niklas Vaara, Lam Huynh, Pekka Sangi, Miguel Bordallo López, Janne Heikkilä |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 60 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.07781