Identification of the Dangerous Meteorological Objects on Doppler-Polarimetric Radar Data Using the Neural Network Based Algorithm. Part 1: Statistical Modeling

This article deals with the process of identification of the dangerous meteorological objects using the neural network based algorithm. Dangerous objects are considered by the example of probable aircraft icing zones. In the first part of this paper theoretical models of microwave backscattering on water drops and ice crystals are considered. The results of statistical calculation on these models are used to train the network. In the second part of this research the identification algorithm will be discussed. Checking of the theoretical calculation of Doppler-polarimetric variables is done on the basis of the experimental data.