On the optimum estimation of 1-minute integrated rainfall statistics from data with longer integration time

The 1-minute integrated cumulative distribution of rain is the most relevant input for the prediction of rain attenuation, the dominant impairment affecting microwave links above approximately 10 GHz. Various techniques exist for the prediction of such a distribution, among them the conversion of statistics with longer integration time. In this study, a comparison of the predictions obtained from statistics conversion models against a database of measurements collected at 34 sites worldwide is conducted, with the objective of determining the degree of usability of the models and of identifying the best performing one. The results show that the semi-physical approach of the EXCELL RSC model provides the overall best performance.