This paper presents the results of Doppler-polarimetric radar signal analysis, computer modeling, and real data processing that show the correlation between differential Doppler velocity (DDV), which is a difference between mean Doppler velocities measured at orthogonal polarizations of electromagnetic waves, and parameters of rain including both rain microstructure (parameters of drop-size distribution) and turbulence intensity in resolution volume. The reasonability to use DDV as one of the informative parameters in the recognition neural network system is checked.