Scalable and Adaptive RF Signal Propagation Modeling for Next-Generation Wireless Systems

Next-generation wireless technologies, such as IEEE 802.11be/bf and Sixth Generation (6G), are designed to support both communication and sensing, enabling a broad range of applications powered by deep learning (DL) models. These models require large-scale, labeled Radio Frequency (RF) datasets, making data collection a bottleneck. Recent Neural Radiance Fields (NeRF)-based RF propagation models offer a promising solution by synthesizing RF signals at arbitrary transceiver positions within complex 3D environments, achieving high-fidelity RF data generation. However, these methods face key limitations in scalability and temporal adaptation, limiting their applicability in real-world settings. This work addresses these challenges by introducing an RF propagation modeling framework that is both scalable and adaptable to dynamic temporal variations, enabling practical next-generation wireless systems.