Precisely modeling radio propagation in complex environments has been a significant challenge, especially with the advent of 5G and beyond networks, where managing massive antenna arrays demands more detailed information. Traditional methods, such as empirical models and ray tracing, often fall short, either due to insufficient details or with challenges for real-time applications. Inspired by the newly proposed 3D Gaussian Splatting method in computer vision domain, which outperforms in reconstructing optical radiance fields, we propose RF-3DGS, a novel approach that enables precise site-specific reconstruction of radio radiance fields from sparse samples. RF-3DGS can render spatial spectra at arbitrary positions within 2 ms following a brief 3-minute training period, effectively identifying dominant propagation paths at these locations. Furthermore, RF-3DGS can provide fine-grained Channel State Information (CSI) of these paths, including the angle of departure and delay. Our experiments, calibrated through real-world measurements, demonstrate that RF-3DGS not only significantly improves rendering quality, training speed, and rendering speed compared to state-of-the-art methods but also holds great potential for supporting wireless communication and advanced applications such as Integrated Sensing and Communication (ISAC).

翻译：在复杂环境中精确建模无线电传播一直是一项重大挑战，尤其是在5G及未来网络时代，管理大规模天线阵列需要更详细的信息。传统方法，如经验模型和射线追踪，往往因细节不足或难以满足实时应用需求而存在局限。受计算机视觉领域新提出的、在重建光学辐射场方面表现优异的3D高斯泼溅方法的启发，我们提出了RF-3DGS，这是一种能够从稀疏样本中精确重建特定场景无线辐射场的新方法。RF-3DGS在短短3分钟的训练后，可在2毫秒内渲染任意位置的空间谱，有效识别这些位置的主要传播路径。此外，RF-3DGS还能提供这些路径的细粒度信道状态信息，包括离开角和时延。我们通过实际测量校准的实验表明，与现有先进方法相比，RF-3DGS不仅在渲染质量、训练速度和渲染速度上均有显著提升，而且在支持无线通信及诸如通感一体化等高级应用方面具有巨大潜力。