Page 709 - The Mechatronics Handbook
P. 709

0066_Frame_C23  Page 17  Wednesday, January 9, 2002  1:53 PM


























































                       FIGURE 23.9  Ideal impulse sampling process: (a) bandlimited continuous-time signal and its Fourier transform,
                       (b) train of impulses and its spectra, (c) sampled signal and its spectra.

                       Impulse Sampling 6–9
                       Consider an idealized impulse-sampling process shown in Fig. 23.9, where x(t) is to be sampled using
                       the train of impulses p(t), so that

                                                         ∞              ∞
                                    x s t() =  xt()pt() =  xt() ∑  δ tnT s ) =  ∑  xnT s )δ tnT s )  (23.34)
                                                                           (
                                                             (
                                                                                 (
                                                                                   –
                                                               –
                                                        n=∞            n=∞
                                                          –
                                                                        –
                       The main difference between x s (t) and x(nT s ) is that the former is essentially a CT signal with zero values
                       except at the integer values of T s , while the latter is a perfectly selected sample of x(t) as a result of impulse
                       sampling.
                         The Fourier transform of Eq. (23.34) yields
                                                                     ∞
                                                                         (
                                             X s f() =  Xf() ∗ Pf() =  F s ∑  Xf F s )          (23.33)
                                                                           –
                                                                    k=∞
                                                                      –
                       where X( f ) is the spectrum of x(t).
                      ©2002 CRC Press LLC
   704   705   706   707   708   709   710   711   712   713   714