OPTIMIZATION IN SOLVING INVERSE PROBLEMS OF ATMOSPHERIC OPTICS BY THE MONTE-CARLO METHOD

The paper considers modifications which allow one to optimize the solution of inverse problems of atmospheric optics. The modifications involve speeding up direct calculations by the Monte Carlo method, using the approximate value functions, and regularizing the corresponding systems of equations by using additional information. Particular attention has been given to two inverse problems, namely the recovery of altitudinal variations of the scattering coefficient and recovery of the scattering indicatrix on the basis of the observations of the angular distribution of scattered solar radiation. The convergence of the algorithm for recovering the scattering indicatrix is investigated theoretically.