RadTwin: Generalizable Wireless Digital Twin for Dynamic Environments

TL;DR

RadTwin is a novel wireless digital twin framework that models radio propagation in dynamic environments without retraining, using scene geometry conditioning and physics-informed neural modules.

Contribution

It introduces a generalizable framework combining scene representation, electromagnetic ray tracing, and neural decoding for adaptive wireless environment modeling.

Findings

RadTwin outperforms NeRF2 with 31.6% higher SSIM and 91.96% lower LPIPS.

It demonstrates high generalization and data efficiency in dynamic indoor scenes.

RadTwin effectively models radio propagation without environment-specific retraining.

Abstract

Precisely modeling radio propagation in dynamic wireless environments is fundamental to the realization of wireless digital twins. Traditional ray tracing methods rely on accurate 3D models with detailed environment parameters, while recent neural radiance field approaches learn representations tied to specific static scenes, requiring retraining when environments change. In this paper, we propose RadTwin, a generalizable wireless digital twin framework that explicitly conditions on scene geometry, enabling adaptation to dynamic environments without retraining. RadTwin comprises three key components: 1) a scenario representation network that extracts high-level latent scene features from point clouds, 2) an electromagnetic ray tracing module that computes physics-informed sparse attention masks identifying voxels that physically contribute signals toward each query direction, and 3) a neural propagation decoder that aggregates relevant scene features through masked cross-attention to learn how radio propagation behaves within the given scene geometry. We evaluate RadTwin on a customized dataset of indoor scenes with varying furniture arrangements. Experimental results show that RadTwin achieves 31.6% higher SSIM (0.846 vs. 0.643) and 91.96% lower LPIPS (0.023 vs. 0.286) compared to NeRF2. RadTwin further demonstrates superior cross-scale performance and high generalization and data efficiency, representing a significant advancement toward practical digital network twins for dynamic wireless environments.