The transport of light through the atmosphere is modeled as a Levy-type Markovian jump process of stochastic corpuscular multiple scattering of photons including polarization. The backward Kolmogorov differential equation of this Markovian jump process is the general radiative transfer equation with polarization. Based on this stochastic multiple scattering process, Monte Carlo codes have been designed to calculate lidar returns containing essential contributions of multiple scattering (dense clouds, space). Since the retrieval of micro-physical cloud parameters is an ill-posed problem, it is necessary to collect as much additional information about the cloud as possible. Such information can be obtained by using multiple field of view lidars or CCD lidars. Based on the stochastic multiple scattering process, a new Monte Carlo code has been designed allowing for the calculation of the off-axis diffusion patterns of the emitted pulsed laser beam as it is "seen from the monostatic or bistatic CCD receiver." These patterns allow for the classification of different types of scattering particles. Some examples of such patterns will be shown for clouds of radiative fog and of collections of randomly oriented oblate and prolate cylinders and spheroids. These patterns allow for a simple classification of clouds.© (2005) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.