Page 553 - Fiber Optic Communications Fund
P. 553

534                                                               Fiber Optic Communications


            where E (r, t) is the slowly varying function of time and c.c stands for complex conjugate. Substituting
                   0
            Eq. (B.9) in Eq. (B.6), we find
                                    (3)
                                   0
                                                                  3
                                               2
                         P (r, t)=     [3|E(r, t)| E(r, t) exp(−i t)+ E (r, t) exp(−i3 t)] + c.c.  (B.10)
                                                                              0
                                                            0
                          NL
                                    8
            The first term in the square bracket corresponds to oscillations at  and the second term corresponds to third
                                                                 0
            harmonic frequency 3 . The efficiency of third harmonic generation in fibers is very small unless special
                               0
            phase-matching techniques are used. Therefore, ignoring the second term and substituting Eqs. (B.7)–(B.10)
            in Eq. (B.1), we obtain
                                 2
                              1  Ψ(r, t)  1  2  (1)          3 (3)   2  2
                     2
                    ∇ Ψ(r, t)−          =      [ (r,t) ⊗ Ψ(r, t)] +   [|Ψ(r, t)| Ψ(r, t)],  (B.11)
                                           2
                              c 2  t 2  c t 2                4c 2  t 2
            where
                                            Ψ(r, t)= E (r, t) exp(−i t),
                                                                0
                                                    0
                                                     1
                                                2
                                               c =     .
                                                    
                                                     0 0
            The electric field intensity in a single-mode fiber may be written as (see Chapter 2)
                                          Ψ(r, t)= q(z, t)(x, y)e −i( 0 t− 0 z) ,      (B.12)
            where  =  ( ) is the propagation constant, (x, y) is the transverse field distribution, and q(z, t) is the
                          0
                   0
                       0
            field envelope which is a slowly varying function of t and z. Substituting Eq. (B.12) in Eq. (B.11) and taking
            the Fourier transform, we obtain
                                    [  2                               ]
                                       ̃q(z, Ω)    ̃q(z, Ω)  2
                                            + 2i      −  ̃q(z, Ω)
                                                   0
                                                               0
                                         z 2         z
                                    {  2     2     2              }
                                              [  (1)  ]
                                  +       +     +     1 + ̃ (r,)  ̃ q(z, Ω)
                                      x 2  y 2  c 2
                                        2
                                     3  (3)  3           ∗
                                  =−        { (x, y)[̃q(z, Ω) ⊗̃q (z, −Ω) ⊗̃q(z, Ω)]},     (B.13)
                                       4c 2
            where Ω=  −  . To obtain Eq. (B.13), we have used the Fourier transform relations
                           0
                                                (  2  )
                                                   A
                                                      =− A()                            (B.14)
                                                            2 ̃
                                                  t 2
            and
                                                        ̃
                                                 ̃
                                      [A(t)B(t)] = A() ⊗ B()
                                                     (  ′  ) (     ′  )
                                                    ̃
                                               =    A    B ̃   −   d .               (B.15)
                                                 ∫     2      2   2
            Under the slowly varying envelope approximation, the first term in Eq. (B.13) can be ignored, which is a good
            approximation for pulse widths that are much longer than the period 2∕ . From Eq. (10.29), we have
                                                                        0
                                                  (1)
                                                            2
                                             1 + ̃ (r,)= n (r,),                       (B.16)
   548   549   550   551   552   553   554   555   556   557   558