Page 276 - Fiber Optic Communications Fund
P. 276

Optical Amplifiers                                                                 257


           From Eq. (6.35), we have
                                               2
                                               =  2 s−sp  +  2 sp−sp .                (6.68)
           Using Eqs. (6.60) and (6.65) in Eq. (6.68), we find

                               2
                                   2
                               = R  ASE o  out  +  ASE o  f ≤ f e
                                                    f ] if
                                        f [2P
                                                            o
                                   2
                                = R   f [2P f +   (2f − f )f ] if  f < f < 2f
                                     ASE o  out o  ASE  o  e e      e   o    e
                                   2
                                = R   [4P f +   (2f − f )f ]  otherwise.                (6.69)
                                     ASE   out e  ASE  o  e e
           When the signal power P  is much larger than the noise power P  , the dominant contribution to the
                                out                               ASE
           variance comes from the signal–ASE beat noise. However, when the noise power is comparable with
           the signal power, the second term in Eq. (6.68) becomes comparable with the first term, as illustrated in
           the following example.
           Example 6.3
           An optical amplifier at 1530 nm has a gain G = 30 dB and n = 5. The input power of the amplifier P is
                                                           sp
                                                                                              in
           −27 dBm. Calculate (a) the variance of the signal–ASE beat noise current, (b) the variance of the ASE–ASE
           beat noise current, and (c) the total variance. Assume R = 0.8 A/W, optical and electrical filters are ideal with
           f = 16 GHz and f = 9 GHz. Repeat this example if P =−60 dBm.
           o
                         e
                                                      in
           Solution:
                                                 c
                                             f =   = 196.08 THz,
                                             0
                                                 
                                      P (dBm)= P (dBm)+ G(dB)
                                       out        in
                                               =−27 + 30 dBm = 3dBm,
                                      P  = 10 P out (dBm)∕10  mW = 2 × 10 −3  W
                                       out
                                            G = 10 G(dB)∕10  = 1000.

           From Eq. (6.17), we have
                                ASE  = n hf (G − 1)
                                       sp
                                          0
                                                                12
                                    = 5 × 6.626 × 10 −34  × 196.08 × 10 ×(1000 − 1)
                                    = 6.48 × 10 −16  W/Hz.
           (a) Using Eq. (6.60), we find
                                                      2
                                              2 s−sp  = 4R P   B ,
                                                        out ASE eff
                                         B   = min{f ∕2, f }= 8 GHz,
                                          eff      o   e
                                2          2          −16      −3        9  2
                                  = 4 × 0.8 × 6.48 × 10  × 2 × 10  × 8 × 10 A             (6.70)
                                s−sp
                                                  2
                                    = 2.657 × 10 −8  A .                                    (6.71)
   271   272   273   274   275   276   277   278   279   280   281