Physics-Informed Generative Modeling of Wireless Channels

Learning the distribution of the wireless channel within a specific environment of interest is essential to exploit the full potential of machine learning (ML) for wireless communications and radar applications. Generative modeling offers a promising framework to address this problem. However, existing approaches pose unresolved challenges, including the need for high-quality training data, limited generalizability, and a lack of physical interpretability. To address these issues, we propose a model that combines the physics-related compressibility of wireless channels with sparse Bayesian generative modeling (SBGM) to learn the distribution of the underlying physical channel parameters. By leveraging the sparsity-inducing characteristics of SBGM, our method can learn from compressed observations received by an access point (AP) during default online operation. Moreover, it is physically interpretable and generalizes to arbitrary system configurations without requiring retraining.