A framework for minimizing the impact of wet antenna attenuation on rainfall estimates provided by commercial microwave links

The compensation of the extra attenuation introduced by wet antennas is crucial to get reliable rainfall intensity estimates from the data collected by commercial microwave links (CMLs). This study reviews the available wet antenna attenuation (WAA) models and proposes a new framework for calibrating WAA model parameters, based on rain gauge measurements (assumed as ground truth). The framework has been applied to the data measured by 86 CMLs located in the Seveso River basin (Northern Italy) during 44 rainy days in 2019-2020. As most WAA models are basically equivalent in spite of different formulations, we selected only a subset, including the Schleiss-Rieckerman-Berne model (SRB) and the Valtr-Fencl-Bareš model, the latter in the versions with the original parameter values (VFB) and with the parameters calibrated over our CML data (VFBm). We found that WAA increases with rainfall intensity, as predicted by the VFB model, and it is weakly dependent on link frequency. We derived two sets of optimum WAA model parameters for the Ka-band (17-23 GHz) and the Q-band (37-43 GHz), respectively. WAA values are up to 4 dB at heavy rainfall intensities (50 mm/h) in the Q-band. The VFBm model performs better than VFB and outperforms SRB, when assessed over the entire set of CML data and over different classes of CMLs (based on distance to the neighbor rain gauge, frequency and path length) or, again, over different classes of rainfall intensity.