From: luweitest@address.invalid   
      
   On 2016-8-19 22:26, Phil Hobbs wrote:   
   > On 08/19/2016 03:19 AM, Lu Wei wrote:   
   >> I am trying to solve a general but seems complicated problem: reflection   
   >> and transmission of light, on an interface of general profile. For   
   >> example, the light falls on a fuzzy glass. Fresnel's equation gives   
   >> analytical results on flat surface, but on rough surface, how do I   
   >> calculate the reflection and transmission? Is there any software that   
   >> could do this?   
   >>   
   >   
   > Depends how rough.  If the height of the roughness is much smaller than   
   > a wavelength, you can do it easily by Fourier optics--it's just a random   
   > phase screen, and the spatial frequency spectrum of the scatter is the   
   > same as that of the surface.  (There's a factor of 2 for reflection   
   > since the phase difference is twice as large.)   
      
   Would you show me how to do this? For a simplified case, if the surface   
   is like regular hills: ^^^^^^^^^^^^^^ How do I calculate the R and T?   
   (assuming wavelength=λ, RI=n, conic hill height= h, width=w, distance=a)   
   >   
   > If the surface is really rough, so that multiple scattering is   
   > important, you pretty well have to do it numerically.  Breault Research   
   > sells ASAP which can do stuff like that.   
   >   
   > In the middle, there are probably physical-optics things you can do to   
   > make a semi-analytical model.   
      
   The hill size in fact could be 100nm~5μm, seems neither small nor big.   
   But the size will not be very different in one calculation, and my   
   priority is the regular case, e.g. hills are all the same size and   
   distributed at the same distance on the surface.   
      
   --   
   Regards,   
   Lu Wei   
   PGP key ID: 0x A12F EF75 92CC E1EA   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)