Page 378 - Schaum's Outline of Theory and Problems of Advanced Calculus
P. 378

CHAP. 14]                       FOURIER INTEGRALS                               369


                              This is equivalent to
                                                       a         1  2 sin  a
                                                      ð         ð      2
                                                           2
                                                         ð1Þ dx ¼      2  d
                                                        a        1
                                                   ð     2       ð    2
                                                    1  sin  a     1  sin  a
                           or                               d  ¼ 2       d  ¼  a
                                                          2            2
                                                     1            0
                                                          ð    2
                                                           1  sin  a   a
                           i.e.,                               2  d  ¼
                                                           0           2
                              By letting  a ¼ u and using Problem 14.5, it is seen that this is correct. The method can also be used to
                              ð   2
                               1  sin u
                           find    2  du directly.
                               0  u
                     PROOF OF THE FOURIER INTEGRAL THEOREM
                     14.8. Present a heuristic demonstration of Fourier’s integral theorem by use of a limiting form of
                           Fourier series.
                              Let
                                                            1
                                                           X       n x      n x
                                                       a 0
                                                              a n cos  þ b n sin
                                                  f ðxÞ¼  þ                                           ð1Þ
                                                        2           L        L
                                                           n¼1
                                    1  ð L    n u            1  ð  L  n u
                                        f ðuÞ cos  du and        f ðuÞ sin  du:
                           where a n ¼                   b n ¼
                                    L  L       L             L  L      L
                              Then by substitution (see Problem 13.21, Chapter 13),
                                                   1  ð  L     1  X  L     n
                                                                 1 ð
                                                                      f ðuÞ cos  ðu   xÞ du
                                              f ðxÞ¼    f ðuÞ du þ  L       L                         ð2Þ
                                                   2L  L            L
                                                                n¼1
                                           ð
                                            1
                              If we assume that  j f ðuÞj du converges, the first term on the right of (2) approaches zero as L !1,
                                             1
                           while the remaining part appears to approach
                                                        1  X  1      n
                                                          1 ð
                                                     lim       f ðuÞ cos  ðu   xÞ du
                                                    L!1 L            L                                ð3Þ
                                                         n¼1   1
                           This last step is not rigorous and makes the demonstration heuristic.
                              Calling    ¼  =L,(3)can be written
                                                                 1
                                                                X
                                                       f ðxÞ¼ lim     Fðn   Þ                         ð4Þ
                                                              !0
                                                                n¼1
                           where we have written
                                                           1  ð  1
                                                               f ðuÞ cos  ðu   xÞ du
                                                     Fð Þ¼     1                                      ð5Þ
                              But the limit (4)isequal to
                                                   ð           ð   ð
                                                    1        1  1   1
                                                                 d    f ðuÞ cos  ðu   xÞ du
                                              f ðxÞ¼  Fð Þ d  ¼
                                                    0          0     1
                           which is Fourier’s integral formula.
                              This demonstration serves only to provide a possible result. To be rigorous, we start with the integral
                                                      1  ð 1  ð  1
                                                         d     f ðuÞ cos  ðu   xÞ dx
                                                        0
                                                             1
                           and examine the convergence.  This method is considered in Problems 14.9 through 14.12.
   373   374   375   376   377   378   379   380   381   382   383