An iterative solution is illustrated of the three-dimensional radiative transfer equation for a horizontally finite and vertically inhomogeneous precipitating cloud. The method is applied to modeling a convective rain cell of cylindrical shape, characterized by spherical raindrops having a negative-exponential drop size distribution. The realistic model also takes into account the presence of a cloud and an ice layer above the rain cell itself. The simulated brightness temperature, the mean radiative temperature, and the path attenuation are evaluated in a three-dimensional geometry from a surface observation point in order to simulate a ground-based station with a beacon receiver and a multichannel radiometer. Numerical results are shown to illustrate the potential of the proposed model for different sets of frequency channels, observation geometries, cloud sizes and types, and precipitation intensities. After generating a large data set by varying the relevant rain cell parameters, regression analysis is applied to derive a statistical estimation of the total path attenuation from surface rain rate and ground-based radiometric measurements together with the frequency scaling factors for cumuliform clouds in the 10- to 50-GHz band.