# Differentiable High-Performance Ray Tracing-Based Simulation of Radio Propagation with Point Clouds

arXiv: https://arxiv.org/abs/2507.04021 · HTML: https://mlsanity.com/paper/2507.04021/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 30 references that are in the map.
- First posted on arXiv 5 Jul 2025.

## Builds on (most cited first, 9 of 9)

- [The Replica Dataset: A Digital Replica of Indoor Spaces](https://mlsanity.com/paper/1906.05797/) (2019)
- [Mask3D: Mask Transformer for 3D Semantic Instance Segmentation](https://mlsanity.com/paper/2210.03105/) (2022)
- [OneFormer3D: One Transformer for Unified Point Cloud Segmentation](https://mlsanity.com/paper/2311.14405/) (2023)
- [Sionna RT: Differentiable Ray Tracing for Radio Propagation Modeling](https://mlsanity.com/paper/2303.11103/) (2023)
- [Real-Time Digital Twins: Vision and Research Directions for 6G and Beyond](https://mlsanity.com/paper/2301.11283/) (2023)
- [Learning Radio Environments by Differentiable Ray Tracing](https://mlsanity.com/paper/2311.18558/) (2023)
- [Learnable Wireless Digital Twins: Reconstructing Electromagnetic Field with Neural Representations](https://mlsanity.com/paper/2409.02564/) (2024)
- [Ray-Optics Simulations of Outdoor-to-Indoor Multipath Channels at 4 and 14 GHz](https://mlsanity.com/paper/2210.03159/) (2022)
- [Ray Launching-Based Computation of Exact Paths with Noisy Dense Point Clouds](https://mlsanity.com/paper/2403.06648/) (2024)

## Built on by (most cited first, 3 of 3)

- [Vision-Language-Model-Guided Differentiable Ray Tracing for Fast and Accurate Multi-Material RF Parameter Estimation](https://mlsanity.com/paper/2601.18242/) (2026)
- [Differentiable Ray Tracing with Gaussians for Unified Radio Propagation Simulation and View Synthesis](https://mlsanity.com/paper/2605.07781/) (2026)
- [Rethinking Radiomap Blind Prediction with Limited Environment and Configuration Representations](https://mlsanity.com/paper/2609.11255/) (2026)

## Abstract (arXiv)

Ray tracing is a widely used deterministic method for radio propagation simulations, capable of producing physically accurate multipath components. The accuracy depends on the quality of the environment model and its electromagnetic properties. Recent advances in computer vision and machine learning have made it possible to reconstruct detailed environment models augmented with semantic segmentation labels. In this letter, we propose a differentiable ray tracing-based radio propagation simulator that operates directly on point clouds. We showcase the efficiency of our method by simulating multi-bounce propagation paths with up to five interactions with specular reflections and diffuse scattering in two indoor scenarios, each completing in less than 90 ms. Lastly, we demonstrate how the differentiability of electromagnetic computations can be combined with segmentation labels to learn the electromagnetic properties of the environment.