arXiv 2026

Differentiable Ray Tracing with Gaussians for Unified Radio Propagation Simulation and View Synthesis

Abstract

Explicit neural representations such as 3D Gaussian Splatting (3DGS) enable high-fidelity and real-time novel view synthesis, yet optimize for alpha-composited optical appearance rather than ray-intersectable geometry. In contrast, radio-frequency (RF) digital twins require deterministic multi-bounce paths, where the geometry dictates trajectories and their associated attenuation and delay. We introduce a framework enabling differentiable RF propagation simulation directly within visually reconstructed neural scenes, allowing point-to-point path computation between arbitrary 3D locations while preserving high-quality visual rendering. Unlike conventional RF simulation pipelines that rely on manually constructed meshes, we embed Gaussian primitives into a hardware-accelerated ray tracing structure as the underlying spatial representation. By extracting physically meaningful channel impulse responses from visual-only reconstructions, we provide cross-modal evidence that neural reconstructions can serve as unified spatial representations for both electromagnetic propagation simulation and photorealistic view synthesis.

Overview

Method overview

Reconstructed Scenes

RF render CWC render
MMS render Saalasti render

Novel View Synthesis

Novel view synthesis results

Channel Impulse Responses

Power delay profiles extracted from visually reconstructed Gaussian scenes, compared against reference simulators and measurements.

All PDPs

Ours PDP
Ours		 Sionna PDP
Sionna		 Nimbus PDP
Nimbus

BibTeX

@article{vaara2026differentiable,
  title={Differentiable Ray Tracing with Gaussians for Unified Radio Propagation Simulation and View Synthesis},
  author={Vaara, Niklas and Huynh, Lam and Sangi, Pekka and L{\'o}pez, Miguel Bordallo and Heikkil{\"a}, Janne},
  journal={arXiv preprint arXiv:2605.07781},
  year={2026}
}