Wideband RF Radiance Field Modeling Using Frequency-embedded 3D Gaussian Splatting
Summary: arXiv:2505.20714v3 Announce Type: replace-cross
Abstract
Indoor environments typically contain diverse RF signals distributed across multiple frequency bands, including NB-IoT, Wi-Fi, and millimeter-wave. Consequently, wideband RF modeling is essential for practical applications such as joint deployment of heterogeneous RF systems, cross-band communication, and distributed RF sensing. Although 3D Gaussian Splatting (3DGS) techniques effectively reconstruct RF radiance fields at a single frequency, they cannot model fields at arbitrary or unknown frequencies across a wide range.
Introduction
In this paper, we present a novel 3DGS algorithm for unified wideband RF radiance field modeling. RF wave propagation depends on signal frequency and the 3D spatial environment, including geometry and material electromagnetic (EM) properties. To address these factors, we introduce a frequency-embedded EM feature network that utilizes 3D Gaussian spheres at each spatial location to learn the relationship between frequency and transmission characteristics, such as attenuation and radiance intensity.
Methodology
Our approach leverages a dataset containing sparse frequency samples in a specific 3D environment, allowing our model to efficiently reconstruct RF radiance fields at arbitrary and unseen frequencies. The frequency-embedded EM feature network is designed to capture the intricate relationships between the frequency of RF signals and their propagation behavior within the indoor environment.
Dataset and Evaluation
To assess our approach, we introduce a large-scale power angular spectrum (PAS) dataset with 50,000 samples spanning 1 to 94 GHz across six indoor environments. This dataset serves as a benchmark for evaluating the model’s performance and its capability to generalize across different frequencies.
Results
Experimental results show that the proposed model trained on multiple frequencies achieves a Structural Similarity Index Measure (SSIM) of 0.922 for PAS reconstruction. This performance surpasses state-of-the-art single-frequency 3DGS models, which achieve an SSIM of 0.863. The improved SSIM indicates that our model can effectively capture the nuances of RF radiance fields across a wide spectrum of frequencies.
Conclusion
In conclusion, the novel 3DGS algorithm presented in this study offers significant advancements in the modeling of wideband RF radiance fields. By integrating a frequency-embedded EM feature network, our approach overcomes limitations associated with single-frequency models, thereby enhancing the potential for applications in heterogeneous RF environments. Future work will focus on further refining the model and exploring its applicability in more complex scenarios.
Key Takeaways
- Wideband RF modeling is crucial for modern indoor environments with diverse RF signals.
- 3D Gaussian Splatting techniques are enhanced with a frequency-embedded EM feature network.
- A new dataset with 50,000 samples enables evaluation across a wide frequency range.
- Results demonstrate superior performance compared to traditional single-frequency models.