Page 297 - Electrical Engineering Dictionary
P. 297

where H(f ) is the Hilbert transform of f .  engineering applications have a Fourier se-
                              See Fourier amplitude, Fourier phase, Hilbert  ries.
                              transform.
                                                                     Fourier transform  a linear mathematical
                              Fourier plane  a plane in an optical system  transform from the domain of time or space
                              where the exact Fourier transform of an input  functions to the frequency domain. The dis-
                              image is generated.                    crete version of the transform (DFT) can be
                                                                     implemented with a particularly efficient al-
                              Fourier series  Let f(t) be a continuous  gorithm (fast Fourier transform or FFT). The
                              time periodic signal with fundamental period  discrete Fourier transform of a digital image
                              T such that                            represents the image as a linear combination
                                                                     of complex exponentials.
                                        ∞
                                        X      jω 0 t                  The Fourier transform of a continuous
                                f(t) =      c n e  , −∞ <t < ∞
                                                                     time period signal f(t) is given by
                                       n=−∞
                                                                            Z
                                                                              ∞
                              where ω 0 is the fundamental frequency, c 0           −jωt
                                                                     F(ω) =     f(t)e   dt, −∞ <ω < ∞.
                              is a real number, and c n ,n 6= 0 are com-     −∞
                              plex numbers. This representation of f(t) is
                                                                     If the signal f(t) is absolutely integrable and
                              called the exponential Fourier series of f(t).
                                                                     is well behaved, then its Fourier transform
                              The coefficients c n are called the Fourier co-
                                                                     exists. For example, the rectangular pulse
                              efficients and are given by
                                                                     signal
                                     1  Z     −jω 0 t
                                c n =     f(t)e   dt, n = 0, ±1,                    1, − T  ≤ t<  T
                                     T  T                                   f(t) =     2       2
                                     ±2,...                                         0, otherwise
                                                                                              2   ωT
                              The function f(t) can also be expressed as  has Fourier transform F(w) =  ω  sin  2  .
                                                                       The inverse Fourier transform of a signal
                                        ∞                            is given by
                                       X
                              f(t) = c 0 +  (a n cos(nω 0 t)+b n sin(nω 0 t))
                                                                                    Z
                                                                                 1    ∞
                                       n=1                                                   jωt
                                                                          f(t) =        F(ω)e   dω
                                                                                 2π  −∞
                              with Fourier coefficients a n and b n deter-
                              mined from                             See also optical Fourier transform, two-di-
                                                                     mensional Fourier transform.
                                       Z
                                     2
                                a n =     f(t) cos(nω 0 tdt), n = 1,
                                     T  T                            Fourier, Jean Baptiste Joseph  (1768–
                                     2, 3,...                        1830) Born: Auxerre, France
                                       Z
                                     2                                 Fourier is best known for the development
                                b n =     f(t) sin(nω 0 tdt), n = 1,
                                     T  T                            of new mathematical tools. The Fourier se-
                                     2, 3,...                        ries, which describes complex periodic func-
                                                                     tions, andtheFourierintegraltheorem, which
                              This representation of f(t) is called the  allows complex equations to be broken into
                              trigonometric Fourier series of f(t). A sig-  simplertrigonometricequationsforeasierso-
                              nal f(t) has a Fourier series if it satisfies  lution, are named in honor of their discoverer.
                              Dirichlet conditions, given by (i) f(t) is ab-  Fourier was an assistant lecturer under the
                              solutely integrable over any period, (ii) f(t)  great mathematicians Joseph Lagrange and
                              is piecewise continuous over any period, and  Gaspard Monge at the Ecole Polytechnique
                              (iii)  d  f(t) is piecewise continuous over any  in Paris. He also served in a number of posi-
                                 dt
                              period. Virtually all periodic signals used in  tions in Napoleon’s government, eventually


                              c 
2000 by CRC Press LLC
   292   293   294   295   296   297   298   299   300   301   302