Page 393 - Fiber Optic Communications Fund
P. 393

374                                                               Fiber Optic Communications


            Using Eqs. (8.290), (8.291), and (8.293), we obtain
                                 ∞          ∞
                       P(0|1)=     p (i )di   p (i )di
                               ∫    1 1  1  ∫  1 0  0
                                0          i 1
                                                           √
                                            [         ]   ⎛      2 ⎞
                                                 2
                                       ∞       RE + i        i RE       ∞    (     )
                                                             1
                                 1               1   1    ⎜      1 ⎟             i 0
                             =           exp −          I          di 1   exp −      di 0
                               (2 ) ∫ 0        2 2  0 ⎜   2  ⎟  ∫         2 2
                                  2 2
                                                          ⎜       ⎟    i 1
                                                          ⎝       ⎠
                                                          √
                                          [          ]  ⎛       2 ⎞
                                               2
                                     ∞       RE + 2i        i RE
                                                            1
                                1              1    1   ⎜      1 ⎟
                             =         exp −           I          di .                       (8.295)
                                                                   1
                               2 ∫ 0         2 2   0 ⎜   2  ⎟
                                 2
                                                        ⎜        ⎟
                                                        ⎝        ⎠
                                2
                ′
            Let i = 2i and x = RE ∕2. Now, Eq. (8.295) becomes
                     1
                1              1
                                                                 √
                                                    [        ]  ⎛   ′ ⎞
                                               ∞            ′       i x
                                          1            2x + i 1  ⎜  1 ⎟  ′
                                  P(0|1)=        exp −         I       di
                                         4 ∫ 0        2 2  0 ⎜ ⎜   2  ⎟  1
                                            2
                                                                      ⎟
                                                                ⎝     ⎠
                                                                     √
                                                         [      ]   ⎛   ′ ⎞
                                                    ∞          ′       i x
                                          1   −  x          x + i 1  ⎜  1 ⎟  ′
                                       =     e  2 2  exp −      I       ⎟  di 1
                                         4 2   ∫ 0        2 2  0 ⎜   2
                                                                    ⎜     ⎟
                                                                    ⎝     ⎠
                                                  ∞
                                             x
                                         1 −
                                       = e   2 2  p (i )di 1
                                                    1 1
                                         2     ∫ 0
                                              (       )
                                         1         E 1
                                       =   exp −        .                                    (8.296)
                                         2       2
                                                   ASE
            For FSK, E = E = E . Using Eq. (8.268), Eq. (8.296) may be rewritten as
                      1
                               av
                          0
                                                         (   DD  )
                                                     1      
                                             P(0|1)=  exp −       .                          (8.297)
                                                     2        2
            Case (ii): bit ‘0’ transmitted. Owing to the symmetry of the problem, P(0|1) is same as P(1|0). The error
            probability is
                                                                 (   DD  )
                                           1                1       
                                      P =   [P(0|1)+ P(1|0)] =  exp −                        (8.298)
                                       b
                                           2                2        2
            Note that this error probability is the same as that given by Eq. (8.251) for asynchronous detection if we
            replace  het  by 2 DD . If we ignore shot noise, we see that two expressions are identical.
            8.5.3   DPSK
            In the case of PSK, the information is transmitted as the absolute phase of the complex field envelope s(t).
            But in the case of DPSK, the information is transmitted as the phase of the field envelope relative to the
            previous bit. To estimate the absolute phase of the transmitted PSK signal, a reference is required at the
            receiver. This reference is provided by the local oscillator whose phase should be synchronized with that of
   388   389   390   391   392   393   394   395   396   397   398