A Geometric Algebra-Informed 3D Gaussian Splatting Framework for Wireless Scene Representation

TL;DR

This paper presents GAI-GS, a novel 3D Gaussian splatting framework using geometric algebra for wireless scene modeling, explicitly capturing ray-object interactions and electromagnetic effects.

Contribution

The paper introduces a geometric algebra-informed neural framework that models wireless propagation effects within a unified 3D scene representation.

Findings

GAI-GS outperforms existing methods on multiple indoor datasets.

It effectively models multipath, attenuation, reflection, and diffraction.

The approach provides a physics-grounded, end-to-end neural solution.

Abstract

In this paper, we introduce Geometric Algebra-Informed 3D Gaussian Splatting (GAI-GS), a framework for wireless modeling that couples 3D Gaussian splatting with a geometric algebra-based attention mechanism to explicitly model ray-object interactions in complex propagation environments. GAI-GS encodes joint spatial-electromagnetic (EM) relations into token representations, enabling scene-level aggregation within a unified, end-to-end neural architecture. This design grounds wireless ray propagation in electromagnetic principles, allowing token interactions to model key effects such as multipath, attenuation, and reflection/diffraction. Through extensive evaluations on multiple real-world indoor datasets, GAI-GS consistently surpasses current baselines across various wireless tasks.