The answer depends strongly on the structures you are interested in. If we talk about layer stacks where you can safely assume quasi-infinitness in the directions perpendicular to the stack direction I use CODE (https://wtheiss.com/wordpress/?cat=2 ). If you want freeware, RefFIT would be an alternative (https://reffit.ch/ ). Both can solve inverse problems. If the layers are structured and periodic, you could use numerical methods like Finite Difference Time Domain (FDTD) or Finite Element Method modelling (FEM) or Rigorous Coupled Wave Analysis (RCWA). It depends somewhat on your problem at hand, but I usually use FDTD for such problems, but they are limited to direct problems. Can you tell us the concrete problem you are trying to solve?
As Thomas Mayerhöfer correctly said, it depends on the problem which you want to solve. In our lab we used Comsol Multiphysics (it solves the problem using Finite Element Method modelling) for calculating absorption, transmission and propagation of light in ice volume. We chose this one because you can build any geometry that you want (In our case, inside of ice volume we putted inclusions with different characteristics as size, forms, dielectric properties, etc.). Unfortunately, this program is not free and requires high computer power.
Hello Svetlana Malashevich, can you please provide doi or title of your paper (if exists) where you used Comsol MultiPhy for modeling light transmission through ice? I am very interested in this topic. Thank you.
Jaroslav Kastyl, the article currently in progress. It will be about the volumetric scattering model in sea ice using the Monte Carlo method. For this Comsol we use more like auxiliary instrument (optional). I can send you some outputs of the Comsol calculations that we have done.