Page 333 - Introduction to Information Optics
P. 333

3 1 8                   6. Interconnection with Optics

          The boundary condition in Eq. (6.9) can be written as

                            D'(h) + (U h - P)D(h) = Q(h)S
                                                                      (6.19)
                            D'(0) +(f/ 0 -P)D(0) -0.

        Using Eqs. (6.17) and (6.19) we can find constant matrices A, B:

       A = (W l + U h - P)~\W 2 + U h - P)B
       B=[_(W 2 + U h-P)Qxp(W 2h)
                    w
                          U
                                                                      l
                 - ( i + H - P}Z*P(WMW, + u  - pr'(W  + u  - P)T Q s
                                                h
                                                         2
                                                                        h
                                                               h
                                                                      (6.20)
          The above equation gives an exact analytical solution in the case of a tilted
       grating. In the case of a rectangular grating, W l and W 2 matrices are

          To calculate exponential function in Eq. (6.20) we can factorize W v and W 2
       matrices:



       where O m is a diagonal matrix the elements of which are the eigen-values of
       W m, and Z m is an orthogonal matrix whose columns are eigen-vectors of W m.
       Using exponential function series and matrix factorization we can write:

                       00     k          1     k k
                         (W h\
                                   °° (7~ W7 } h
                            n
                          vv
                       — \ m >    V^ V-^m  vv m^m)  n  _ y-
                                   \       —      — Z< m
                           Jc!
                                                                    1
          Thus, the exponential function in Eq. (6.20) is: exp(W mh) = Z m O mZ OT,
       where O m is a diagonal matrix the elements of which are [Q] n = exp(Q, nh).
       Coefficients R n and T n in Eqs. (6.3), (6.4) can be found from condition of E,
       being continued at y — h and y = 0. Energy in the nth diifracted, reflected, and
       transmitted orders can be found


                                 E R = R*R ncos<x nl
                                 E T= T*T ncosx n2
       where * means complex conjugate.
   328   329   330   331   332   333   334   335   336   337   338