Bayesian Rain Field Reconstruction using Commercial Microwave Links and Diffusion Model Priors
For researchers in environmental sensing and inverse problems, this provides a training-free Bayesian framework that better preserves rainfall statistics than censored Gaussian processes.
This work formulates rain field reconstruction from Commercial Microwave Links as a Bayesian inverse problem using diffusion models as spatial priors, achieving consistent improvements over existing baselines on synthetic and real-world datasets.
Commercial Microwave Links (CMLs) offer dense spatial coverage for rainfall sensing but produce path-integrated measurements that make accurate ground-level reconstruction challenging. Existing methods typically oversimplify CMLs as point sensors and neglect line integration relating rainfall to signal attenuation, resulting in degraded performance under heterogeneous precipitation. In this work, we view rain field reconstruction as a Bayesian inverse problem with Diffusion Models (DMs) as high-fidelity spatial priors. We show that diffusion models better preserve key rainfall statistics compared to censored Gaussian processes. Framing rainfall estimation as a Bayesian inverse problem with a DM prior enables training-free posterior sampling using a broad family of methods, including Plug-and-Play, Sequential Monte Carlo, and Replica Exchange methods. Experiments on synthetic and real-world datasets demonstrate consistent improvements over established CML-based reconstruction baselines.