图14.3富里哀级数的结果比较 注意上述两个富里哀级数给出了相似的结果。从图143中可知,在锯齿信号的不连续 点的周围出现了明显的吉布现象波纹。用函数 window,将一个窗口加于富里哀级数系数, 就能减少这种波纹 >>help fswindow FSWINDOW Generate Window Functions FSWINDOW(N, TYPE)creates a window vector of type TYPE having FSWINDOW(X, TYPE)creates a window vector of type TYPE having a length and orientation the same as the vector X FSWINDOW(X, TYPE, alpha) porvides a parameter alpha as required for some window type FSWINDOW with no input arguments returns a string matrix whose i-th row is the i-th TYPE given below TYPE is a string designating the window type desired rec= Re lar or box 'tri'= Triangular or bartlet han'= Hann or hanning ham'= hamming ' bla= blackman common coefs 'blx'= blackman exact coefs rie=Rieman (sin(x)/x) tuk=Tukey, O<alpha<1 alpha=0.5 is default poi'=Poisson, O<alpha<inf; alpha=1 is default cau=Cauchy, 1<alpha<inf; alpha= I is default gau'=Gaussian, I<alpha<inf; alpha= I is default Reference: F J. Harris, On the Use of windows for Harmonic Analysis with the Discrete ourier Transform, " IEEE Proc. Vol. 66, no. 1, Jan. 1978, pp51-83 >>Fnh=Fn. 'fswindow(Fn, han); apply Hanning window >>f-fseval( Fnh, t, wo) evaluate windowed FS >>plot(t, f) plot results图 14.3 富里哀级数的结果比较 注意上述两个富里哀级数给出了相似的结果。从图 14.3 中可知，在锯齿信号的不连续 点的周围出现了明显的吉布现象波纹。用函数 fswindow，将一个窗口加于富里哀级数系数， 就能减少这种波纹。 >>help fswindow FSWINDOW Generate Window Functions. FSWINDOW(N, TYPE) creates a window vector of type TYPE having a length equal to the scale N. FSWINDOW(X, TYPE) creates a window vector of type TYPE having a length and orientation the same as the vector X. FSWINDOW(X, TYPE, alpha) porvides a parameter alpha as required for some window types. FSWINDOW with no input arguments returns a string matrix whose i-th row is the i-th TYPE given below. TYPE is a string designating the window type desired: 'rec' = Rectangular or Boxcar 'tri' = Triangular or Bartlet 'han' = Hann or hanning 'ham' = Hamming 'bla' = blackman common coefs. 'blx' = Blackman exact coefs. 'rie' = Rieman {sin(x)/x} 'tuk' = Tukey, 0<alpha<1; alpha = 0.5 is default 'poi' = Poisson, 0<alpha<inf; alpha = 1 is default 'cau' = Cauchy, 1<alpha<inf; alpha = 1 is default 'gau' = Gaussian, 1<alpha<inf; alpha = 1 is default Reference: F. J. Harris,”On the Use of Windows for Harmonic Analysis with the Discrete Fourier Transform,” IEEE Proc. Vol. 66, no. 1, Jan. 1978, pp.51-83 >>Fnh=Fn.*fswindow(Fn, ‘ han ‘); % apply Hanning window >>f=fseval(Fnh, t, wo); % evaluate windowed FS >>plot(t, f) % plot results