RxGS: Receiver-Generalizable 3D Gaussian Splatting for Radio-Frequency Data Synthesis

2026-05-22

AI Summary

Existing RF signal synthesis methods require training a separate model for each receiver, suffering from poor generalization to unseen locations and incurring high training and inference costs as well as substantial storage overhead. This work proposes a two-stage, single-model architecture based on 3D Gaussian Splatting (3DGS): the first stage learns a receiver-agnostic shared 3D geometry, while the second stage, with the geometry frozen, jointly models receiver-position-dependent directional radiation through global and local conditioning branches. To enhance efficiency, we also design a multi-receiver batched CUDA rasterizer. Our approach is the first to enable a single model to generalize to arbitrary receiver positions—including unseen ones—matching or surpassing per-receiver baselines across multiple RF datasets, while reducing training costs by up to 45×, accelerating inference by 7.6×, and significantly lowering storage requirements.

Abstract

Radio-frequency (RF) data synthesis predicts the received signal given transmitter and receiver positions, and is essential for wireless applications. Recent 3D Gaussian Splatting (3DGS)-based methods achieve efficient synthesis at any transmitter but only for a fixed receiver. Therefore, supporting $N$ receivers in one scene requires $N$ independent models and precludes prediction at unseen receivers. We present RxGS, which achieves receiver-generalizable synthesis within a single unified model. Our key insight is that scene geometry is receiver-independent while directional radiance is not: a first stage learns shared 3D Gaussian geometry, and a second stage freezes it and learns directional radiance conditioned on receiver position. A global conditioning branch captures shared receiver-dependent effects across the scene, while a local branch models per-scatterer variations from the receiver's geometry and occlusion. A multi-receiver CUDA rasterizer further batches rendering across all $N$ receivers. Evaluated across various RF datasets, RxGS matches or improves over per-receiver baselines with a single shared model and generalizes to receivers unseen during training within the scene, cutting training cost by up to $45\times$, inference cost by $7.6\times$, and storage by $N\times$.

Problem

• receiver-generalizable
• 3D Gaussian Splatting
• RF data synthesis
• unseen receiver
• multi-receiver

Innovation

• receiver-generalizable
• 3D Gaussian Splatting
• RF data synthesis
• directional radiance
• multi-receiver rendering