Near-ground Precipitation Sensing Using Full-duplex MIMO Base Stations

Near-ground precipitation sensing plays a vital role in meteorology, hydrology, and climate science. Ground-based radars and distributed rain gauges are the workhorses of precipitation sensing, but due to their limited scalability, they provide very coarse near-ground precipitation sensing resolution. This paper proposes a novel approach for near-ground precipitation sensing using communication-based full-duplex Multiple-Input Multiple-Output (MIMO) base stations, thereby enabling highly scalable precipitation sensing. We first derive a MIMO channel to characterize combined scattering from precipitation particles and backscatter from objects in the precipitation footprint. The precipitation model leverages the concept of resolution volume from the weather radar literature and captures different radio configurations to match modern base station architectures across diverse spectrum bands. Using the proposed channel models, we derive a suite of transmit/receive beamformers that optimize precipitation sensing under a diverse set of practical constraints. Finally, we use both simulation-based and measurement-based clutter analysis to quantify the performance of the proposed beamformers. In the process, we identify operational regimes to achieve high precipitation estimation accuracy.