File:Window function (rectangular).png

預覽大小：800 × 359 像素。其他解析度：320 × 144 像素 | 640 × 287 像素 | 1,024 × 460 像素 | 1,280 × 575 像素 | 2,500 × 1,123 像素。

原始檔案（2,500 × 1,123 像素，檔案大小：83 KB，MIME 類型：image/png）

本檔案並非來自中文維基百科，而是來自維基共享資源。請參閱維基共享資源上的詳細描述、討論頁、頁面歷史、日誌。

維基共享資源是一個儲存自由版權作品的項目。您可以向此項目作出貢獻。

Transferred from en.wikipedia to Commons by Tiaguito.

摘要

描述

rectangular window and frequency response

日期

2005年12月17日

來源

自己的作品

作者

Bob K (original version), Olli Niemitalo

授權許可
(重用此檔案)

我，此作品的版權所有人，釋出此作品至公共領域。此授權條款在全世界均適用。
這可能在某些國家不合法，如果是的話：
我授予任何人有權利使用此作品於任何用途，除受法律約束外，不受任何限制。

其他版本

File:Window function and frequency response - Rectangular.svg是本檔案的向量版本。如果品質不低，就應該優先使用該檔案，而非PNG檔案。

File:Window function (rectangular).png → File:Window function and frequency response - Rectangular.svg

更多資訊請參閱Help:SVG/zh。

其他語言

Alemannisch ∙ العربية ∙ беларуская (тарашкевіца) ∙ български ∙ বাংলা ∙ català ∙ нохчийн ∙ čeština ∙ dansk ∙ Deutsch ∙ Ελληνικά ∙ English ∙ British English ∙ Esperanto ∙ español ∙ eesti ∙ euskara ∙ فارسی ∙ suomi ∙ français ∙ Frysk ∙ galego ∙ Alemannisch ∙ עברית ∙ hrvatski ∙ magyar ∙ հայերեն ∙ Bahasa Indonesia ∙ Ido ∙ italiano ∙ 日本語 ∙ ქართული ∙ 한국어 ∙ lietuvių ∙ македонски ∙ മലയാളം ∙ Bahasa Melayu ∙ norsk bokmål ∙ Plattdüütsch ∙ Nederlands ∙ norsk nynorsk ∙ norsk ∙ occitan ∙ polski ∙ prūsiskan ∙ português ∙ português do Brasil ∙ română ∙ русский ∙ sicilianu ∙ Scots ∙ slovenčina ∙ slovenščina ∙ српски / srpski ∙ svenska ∙ தமிழ் ∙ ไทย ∙ Türkçe ∙ татарча / tatarça ∙ українська ∙ vèneto ∙ Tiếng Việt ∙ 中文 ∙ 中文（中国大陆） ∙ 中文（简体） ∙ 中文（繁體） ∙ 中文（马来西亚） ∙ 中文（新加坡） ∙ 中文（臺灣） ∙ 中文（臺灣） ∙ +/−

Source code

The script below generates these .png images:

This script has not been tested in MATLAB. See the individual file histories for the simpler MATLAB scripts that were the basis of this script.

Generation of svg files by minor modification of the script displayed visual artifacts and renderer incompatibilities that could not be easily fixed. The current script fixes the visual artifacts in the png file as a post-processing step. The script generates a semi-transparent grid by taking a weighted average of two images, one with the grid and one without.

N

本PNG 點陣圖使用Octave創作Olli Niemitalo。

Matlab

function plotWindowLayer (w, N, gridded, wname, wspecifier)
 
  M=32;
  k=0:N-1;
  dr = 120;

  H = abs(fft([w zeros(1,(M-1)*N)]));
  H = fftshift(H);
  H = H/max(H);
  H = 20*log10(H);
  H = max(-dr,H);
 
  figure('Position',[1 1 1200 520])
  subplot(1,2,1)
  set(gca,'FontSize',28)
  area(k,w,'FaceColor', [0 1 1],'edgecolor', [1 1 0],'linewidth', 2)
  xlim([0 N-1])
  if (min(w) &gt;= -0.01)
    ylim([0 1.05])
    set(gca,'YTick', [0 : 0.1 : 1])
    ylabel('amplitude','position',[-16 0.525 0])
  else
    ylim([-1 5])
    set(gca,'YTick', [-1 : 1 : 5])
    ylabel('amplitude','position',[-16 2 0])
  endif
  set(gca,'XTick', [0 : 1/8 : 1]*(N-1))
  set(gca,'XTickLabel',[' 0'; ' '; ' '; ' '; ' '; ' '; ' '; ' '; 'N-1'])
  grid(gridded)
  set(gca,'LineWidth',2)
  set(gca,'gridlinestyle','-')
  xlabel('samples')
  if (strcmp (wspecifier, ""))
    title(cstrcat(wname,' window'))
  else
    title(cstrcat(wname,' window (', wspecifier, ')'))
  endif
  set(gca,'Position',[0.08 0.11 0.4 0.8])
  set(gca,'XColor',[1 0 1])
  set(gca,'YColor',[1 0 1])
  
  subplot(1,2,2)
  set(gca,'FontSize',28)
  h = stem(([1:M*N]-1-M*N/2)/M,H,'-');
  set(h,'BaseValue',-dr)
  ylim([-dr 6])
  set(gca,'YTick', [0 : -10 : -dr])
  set(findobj('Type','line'),'Marker','none','Color',[0 1 1])
  xlim([-M*N/2 M*N/2]/M)
  grid(gridded)
  set(findobj('Type','gridline'),'Color',[.871 .49 0])
  set(gca,'LineWidth',2)
  set(gca,'gridlinestyle','-')
  ylabel('decibels')
  xlabel('bins')
  title('Frequency response')
  set(gca,'Position',[0.59 0.11 0.4 0.8])
  set(gca,'XColor',[1 0 1])
  set(gca,'YColor',[1 0 1])

endfunction

function plotWindow (w, wname, wspecifier = "", wfilespecifier = "")

  if (strcmp (wfilespecifier, ""))
    wfilespecifier = wspecifier;
  endif

  N = size(w)(2);
  B = N*sum(w.^2)/sum(w)^2   % noise bandwidth (bins), set N = 4096 to get an accurate estimate
  
  plotWindowLayer(w, N, "on", wname, wspecifier);  % "gridded" = "on"
  print temp1.png -dpng "-S2500,1165"
  close
  plotWindowLayer(w, N, "off", wname, wspecifier);  % "gridded" = "off"
  print temp2.png -dpng "-S2500,1165"
  close
% I'm not sure what's going on here, but it looks like the author might have been able
% to save himself some time by using set(gca,"Layer","top") and set(gca,"Layer","bottom").
  I = imread ("temp1.png");
  J = imread ("temp2.png");
  info = imfinfo ("temp1.png");
  w = info.Width;
  c = 1-(double(I(:,1:w/2,1))+2*double(J(:,1:w/2,1)))/(255*3);
  m = 1-(double(I(:,1:w/2,2))+2*double(J(:,1:w/2,2)))/(255*3);
  y = 1-(double(I(:,1:w/2,3))+2*double(J(:,1:w/2,3)))/(255*3);
  c = ((c != m) | (c != y)).*(c &gt; 0).*(1-m-y);
  I(:,1:w/2,1) = 255*(1-c-m-y + 0*m + 0*y + 0*c);
  I(:,1:w/2,2) = 255*(1-c-m-y + 0*m + 0*y + 0.4*c);
  I(:,1:w/2,3) = 255*(1-c-m-y + 0*m + 0*y + 0.6*c);
  c = 1-(double(I(:,w/2+1:w,1))+2*double(J(:,w/2+1:w,1)))/(255*3);
  m = 1-(double(I(:,w/2+1:w,2))+2*double(J(:,w/2+1:w,2)))/(255*3);
  y = 1-(double(I(:,w/2+1:w,3))+2*double(J(:,w/2+1:w,3)))/(255*3);
  c = ((c != m) | (c != y)).*c;
  I(:,w/2+1:w,1) = 255*(1-c-m-y + 0*m + 0*y + 0.8710*c);
  I(:,w/2+1:w,2) = 255*(1-c-m-y + 0*m + 0*y + 0.49*c);
  I(:,w/2+1:w,3) = 255*(1-c-m-y + 0*m + 0*y + 0*c);
  if (strcmp (wfilespecifier, ""))
    imwrite (I, cstrcat('Window function and frequency response - ', wname, '.png'));
  else
    imwrite (I, cstrcat('Window function and frequency response - ', wname, ' (', wfilespecifier, ').png'));
  endif
  
endfunction

N=128;
k=0:N-1;

w = 0.42 - 0.5*cos(2*pi*k/(N-1)) + 0.08*cos(4*pi*k/(N-1));
plotWindow(w, "Blackman")

w = 0.355768 - 0.487396*cos(2*pi*k/(N-1)) + 0.144232*cos(4*pi*k/(N-1)) -0.012604*cos(6*pi*k/(N-1));
plotWindow(w, "Nuttall", "continuous first derivative")

w = 1 - 1.93*cos(2*pi*k/(N-1)) + 1.29*cos(4*pi*k/(N-1)) -0.388*cos(6*pi*k/(N-1)) +0.032*cos(8*pi*k/(N-1));
plotWindow(w, "Flat top")

w = 1 - 1.93*cos(2*pi*k/(N-1)) + 1.29*cos(4*pi*k/(N-1)) -0.388*cos(6*pi*k/(N-1)) +0.028*cos(8*pi*k/(N-1));
plotWindow(w, "SRS flat top")

w = ones(1,N);
plotWindow(w, "Rectangular")

w = (N/2 - abs([0:N-1]-(N-1)/2))/(N/2);
plotWindow(w, "Triangular")

w = 0.5 - 0.5*cos(2*pi*k/(N-1));
plotWindow(w, "Hann")

w = 0.53836 - 0.46164*cos(2*pi*k/(N-1));
plotWindow(w, "Hamming", "alpha = 0.53836")

alpha = 0.5;
w = ones(1,N);
n = -(N-1)/2 : -alpha*N/2;
L = length(n);
w(1:L) = 0.5*(1+cos(pi*(abs(n)-alpha*N/2)/((1-alpha)*N/2)));
w(N : -1 : N-L+1) = w(1:L);
plotWindow(w, "Tukey", "alpha = 0.5")

w = sin(pi*k/(N-1));
plotWindow(w, "Cosine")

w = sinc(2*k/(N-1)-1);
plotWindow(w, "Lanczos")

w = ((N-1)/2 - abs([0:N-1]-(N-1)/2))/((N-1)/2);
plotWindow(w, "Bartlett")

sigma = 0.4;
w = exp(-0.5*( (k-(N-1)/2)/(sigma*(N-1)/2) ).^2);
plotWindow(w, "Gaussian", "sigma = 0.4")

w = 0.62 -0.48*abs(k/(N-1) -0.5) +0.38*cos(2*pi*(k/(N-1) -0.5));
plotWindow(w, "Bartlett–Hann")

alpha = 2;
w = besseli(0,pi*alpha*sqrt(1-(2*k/(N-1) -1).^2))/besseli(0,pi*alpha);
plotWindow(w, "Kaiser", "alpha = 2")

alpha = 3;
w = besseli(0,pi*alpha*sqrt(1-(2*k/(N-1) -1).^2))/besseli(0,pi*alpha);
plotWindow(w, "Kaiser", "alpha = 3")

tau = N-1;
epsilon = 0.1;
t_cut = tau * (0.5 - epsilon);
T_in = abs(k - 0.5 * tau);
z_exp = ((t_cut - 0.5 * tau) ./ (T_in - t_cut) + (t_cut - 0.5 * tau) ./ (T_in - 0.5 * tau));
sigma =  (T_in &lt; 0.5 * tau) ./ (exp(z_exp) + 1);        
w = 1 * (T_in &lt;= t_cut) + sigma .* (T_in &gt; t_cut);
plotWindow(w, "Planck-taper", "epsilon = 0.1")

w = 0.35875 - 0.48829*cos(2*pi*k/(N-1)) + 0.14128*cos(4*pi*k/(N-1)) -0.01168*cos(6*pi*k/(N-1));
plotWindow(w, "Blackman-Harris")

w = 0.3635819 - 0.4891775*cos(2*pi*k/(N-1)) + 0.1365995*cos(4*pi*k/(N-1)) -0.0106411*cos(6*pi*k/(N-1));
plotWindow(w, "Blackman-Nuttall")

w = 1 - 1.93*cos(2*pi*k/(N-1)) + 1.29*cos(4*pi*k/(N-1)) -0.388*cos(6*pi*k/(N-1)) +0.032*cos(8*pi*k/(N-1));
plotWindow(w, "Flat top")

tau = (N/2);
w = exp(-abs(k-(N-1)/2)/tau);
plotWindow(w, "Exponential", "tau = N/2", "half window decay")

tau = (N/2)/(60/8.69);
w = exp(-abs(k-(N-1)/2)/tau);
plotWindow(w, "Exponential", "tau = (N/2)/(60/8.69)", "60dB decay")

alpha = 2;
w = 1/2*(1 - cos(2*pi*k/(N-1))).*exp(alpha*abs(N-2*k-1)/(1-N));
plotWindow(w, "Hann-Poisson", "alpha = 2")

原始碼

Octave

Source code

function plotWindowLayer (w, N, gridded, wname, wspecifier)
 
  M=32;
  k=0:N-1;
  dr = 120;

  H = abs(fft([w zeros(1,(M-1)*N)]));
  H = fftshift(H);
  H = H/max(H);
  H = 20*log10(H);
  H = max(-dr,H);
 
  figure('Position',[1 1 1200 520])
  subplot(1,2,1)
  set(gca,'FontSize',28)
  area(k,w,'FaceColor', [0 1 1],'edgecolor', [1 1 0],'linewidth', 2)
  xlim([0 N-1])
  if (min(w) &gt;= -0.01)
    ylim([0 1.05])
    set(gca,'YTick', [0 : 0.1 : 1])
    ylabel('amplitude','position',[-16 0.525 0])
  else
    ylim([-1 5])
    set(gca,'YTick', [-1 : 1 : 5])
    ylabel('amplitude','position',[-16 2 0])
  endif
  set(gca,'XTick', [0 : 1/8 : 1]*(N-1))
  set(gca,'XTickLabel',[' 0'; ' '; ' '; ' '; ' '; ' '; ' '; ' '; 'N-1'])
  grid(gridded)
  set(gca,'LineWidth',2)
  set(gca,'gridlinestyle','-')
  xlabel('samples')
  if (strcmp (wspecifier, ""))
    title(cstrcat(wname,' window'))
  else
    title(cstrcat(wname,' window (', wspecifier, ')'))
  endif
  set(gca,'Position',[0.08 0.11 0.4 0.8])
  set(gca,'XColor',[1 0 1])
  set(gca,'YColor',[1 0 1])
  
  subplot(1,2,2)
  set(gca,'FontSize',28)
  h = stem(([1:M*N]-1-M*N/2)/M,H,'-');
  set(h,'BaseValue',-dr)
  ylim([-dr 6])
  set(gca,'YTick', [0 : -10 : -dr])
  set(findobj('Type','line'),'Marker','none','Color',[0 1 1])
  xlim([-M*N/2 M*N/2]/M)
  grid(gridded)
  set(findobj('Type','gridline'),'Color',[.871 .49 0])
  set(gca,'LineWidth',2)
  set(gca,'gridlinestyle','-')
  ylabel('decibels')
  xlabel('bins')
  title('Frequency response')
  set(gca,'Position',[0.59 0.11 0.4 0.8])
  set(gca,'XColor',[1 0 1])
  set(gca,'YColor',[1 0 1])

endfunction

function plotWindow (w, wname, wspecifier = "", wfilespecifier = "")

  if (strcmp (wfilespecifier, ""))
    wfilespecifier = wspecifier;
  endif

  N = size(w)(2);
  B = N*sum(w.^2)/sum(w)^2   % noise bandwidth (bins), set N = 4096 to get an accurate estimate
  
  plotWindowLayer(w, N, "on", wname, wspecifier);  % "gridded" = "on"
  print temp1.png -dpng "-S2500,1165"
  close
  plotWindowLayer(w, N, "off", wname, wspecifier);  % "gridded" = "off"
  print temp2.png -dpng "-S2500,1165"
  close
% I'm not sure what's going on here, but it looks like the author might have been able
% to save himself some time by using set(gca,"Layer","top") and set(gca,"Layer","bottom").
  I = imread ("temp1.png");
  J = imread ("temp2.png");
  info = imfinfo ("temp1.png");
  w = info.Width;
  c = 1-(double(I(:,1:w/2,1))+2*double(J(:,1:w/2,1)))/(255*3);
  m = 1-(double(I(:,1:w/2,2))+2*double(J(:,1:w/2,2)))/(255*3);
  y = 1-(double(I(:,1:w/2,3))+2*double(J(:,1:w/2,3)))/(255*3);
  c = ((c != m) | (c != y)).*(c &gt; 0).*(1-m-y);
  I(:,1:w/2,1) = 255*(1-c-m-y + 0*m + 0*y + 0*c);
  I(:,1:w/2,2) = 255*(1-c-m-y + 0*m + 0*y + 0.4*c);
  I(:,1:w/2,3) = 255*(1-c-m-y + 0*m + 0*y + 0.6*c);
  c = 1-(double(I(:,w/2+1:w,1))+2*double(J(:,w/2+1:w,1)))/(255*3);
  m = 1-(double(I(:,w/2+1:w,2))+2*double(J(:,w/2+1:w,2)))/(255*3);
  y = 1-(double(I(:,w/2+1:w,3))+2*double(J(:,w/2+1:w,3)))/(255*3);
  c = ((c != m) | (c != y)).*c;
  I(:,w/2+1:w,1) = 255*(1-c-m-y + 0*m + 0*y + 0.8710*c);
  I(:,w/2+1:w,2) = 255*(1-c-m-y + 0*m + 0*y + 0.49*c);
  I(:,w/2+1:w,3) = 255*(1-c-m-y + 0*m + 0*y + 0*c);
  if (strcmp (wfilespecifier, ""))
    imwrite (I, cstrcat('Window function and frequency response - ', wname, '.png'));
  else
    imwrite (I, cstrcat('Window function and frequency response - ', wname, ' (', wfilespecifier, ').png'));
  endif
  
endfunction

N=128;
k=0:N-1;

w = 0.42 - 0.5*cos(2*pi*k/(N-1)) + 0.08*cos(4*pi*k/(N-1));
plotWindow(w, "Blackman")

w = 0.355768 - 0.487396*cos(2*pi*k/(N-1)) + 0.144232*cos(4*pi*k/(N-1)) -0.012604*cos(6*pi*k/(N-1));
plotWindow(w, "Nuttall", "continuous first derivative")

w = 1 - 1.93*cos(2*pi*k/(N-1)) + 1.29*cos(4*pi*k/(N-1)) -0.388*cos(6*pi*k/(N-1)) +0.032*cos(8*pi*k/(N-1));
plotWindow(w, "Flat top")

w = 1 - 1.93*cos(2*pi*k/(N-1)) + 1.29*cos(4*pi*k/(N-1)) -0.388*cos(6*pi*k/(N-1)) +0.028*cos(8*pi*k/(N-1));
plotWindow(w, "SRS flat top")

w = ones(1,N);
plotWindow(w, "Rectangular")

w = (N/2 - abs([0:N-1]-(N-1)/2))/(N/2);
plotWindow(w, "Triangular")

w = 0.5 - 0.5*cos(2*pi*k/(N-1));
plotWindow(w, "Hann")

w = 0.53836 - 0.46164*cos(2*pi*k/(N-1));
plotWindow(w, "Hamming", "alpha = 0.53836")

alpha = 0.5;
w = ones(1,N);
n = -(N-1)/2 : -alpha*N/2;
L = length(n);
w(1:L) = 0.5*(1+cos(pi*(abs(n)-alpha*N/2)/((1-alpha)*N/2)));
w(N : -1 : N-L+1) = w(1:L);
plotWindow(w, "Tukey", "alpha = 0.5")

w = sin(pi*k/(N-1));
plotWindow(w, "Cosine")

w = sinc(2*k/(N-1)-1);
plotWindow(w, "Lanczos")

w = ((N-1)/2 - abs([0:N-1]-(N-1)/2))/((N-1)/2);
plotWindow(w, "Bartlett")

sigma = 0.4;
w = exp(-0.5*( (k-(N-1)/2)/(sigma*(N-1)/2) ).^2);
plotWindow(w, "Gaussian", "sigma = 0.4")

w = 0.62 -0.48*abs(k/(N-1) -0.5) +0.38*cos(2*pi*(k/(N-1) -0.5));
plotWindow(w, "Bartlett–Hann")

alpha = 2;
w = besseli(0,pi*alpha*sqrt(1-(2*k/(N-1) -1).^2))/besseli(0,pi*alpha);
plotWindow(w, "Kaiser", "alpha = 2")

alpha = 3;
w = besseli(0,pi*alpha*sqrt(1-(2*k/(N-1) -1).^2))/besseli(0,pi*alpha);
plotWindow(w, "Kaiser", "alpha = 3")

tau = N-1;
epsilon = 0.1;
t_cut = tau * (0.5 - epsilon);
T_in = abs(k - 0.5 * tau);
z_exp = ((t_cut - 0.5 * tau) ./ (T_in - t_cut) + (t_cut - 0.5 * tau) ./ (T_in - 0.5 * tau));
sigma =  (T_in &lt; 0.5 * tau) ./ (exp(z_exp) + 1);        
w = 1 * (T_in &lt;= t_cut) + sigma .* (T_in &gt; t_cut);
plotWindow(w, "Planck-taper", "epsilon = 0.1")

w = 0.35875 - 0.48829*cos(2*pi*k/(N-1)) + 0.14128*cos(4*pi*k/(N-1)) -0.01168*cos(6*pi*k/(N-1));
plotWindow(w, "Blackman-Harris")

w = 0.3635819 - 0.4891775*cos(2*pi*k/(N-1)) + 0.1365995*cos(4*pi*k/(N-1)) -0.0106411*cos(6*pi*k/(N-1));
plotWindow(w, "Blackman-Nuttall")

w = 1 - 1.93*cos(2*pi*k/(N-1)) + 1.29*cos(4*pi*k/(N-1)) -0.388*cos(6*pi*k/(N-1)) +0.032*cos(8*pi*k/(N-1));
plotWindow(w, "Flat top")

tau = (N/2);
w = exp(-abs(k-(N-1)/2)/tau);
plotWindow(w, "Exponential", "tau = N/2", "half window decay")

tau = (N/2)/(60/8.69);
w = exp(-abs(k-(N-1)/2)/tau);
plotWindow(w, "Exponential", "tau = (N/2)/(60/8.69)", "60dB decay")

alpha = 2;
w = 1/2*(1 - cos(2*pi*k/(N-1))).*exp(alpha*abs(N-2*k-1)/(1-N));
plotWindow(w, "Hann-Poisson", "alpha = 2")

檔案歷史

點選日期/時間以檢視該時間的檔案版本。

	日期/時間	尺寸	使用者	備⁠註
目前	2013年2月9日 (六) 16:48	2,500 × 1,123（83 KB）	Olli Niemitalo	Antialiasing, layout changes, larger font
	2005年12月17日 (六) 21:07	1,038 × 419（7 KB）	Tiaguito~commonswiki	file size. color source: http://en.wikipedia.org/wiki/Window_Function
	2005年12月17日 (六) 20:48	1,038 × 419（8 KB）	Tiaguito~commonswiki	source: http://en.wikipedia.org/wiki/Window_Function author: http://en.wikipedia.org/wiki/User:Bob_K

檔案用途

沒有使用此檔案的頁面。

全域檔案使用狀況

以下其他 wiki 使用了這個檔案：

da.wikipedia.org 的使用狀況
- Bruger:Peh-herlev/sandkasse-Vinduesfunktion
et.wikipedia.org 的使用狀況
- Kasutaja:U.schults/Aknafunktsioon

File:Window function (rectangular).png

摘要

Matlab

Octave

說明

在此檔案描寫的項目

描繪內容

創作作者

沒有維基數據項目的某些值

著作權狀態

保有知識產權並由其所有者公開於公有領域

授權條款

著作權持有者釋出至公有領域

成立或建立時間

17 12 2005

檔案來源 ^{Chinese (Taiwan) (已轉換拼寫)}

上傳者的原創作品

多媒體型式

image/png

檔案歷史

檔案用途

全域檔案使用狀況

摘要

Matlab

Octave

說明

在此檔案描寫的項目

沒有維基數據項目的某些值

17 12 2005

檔案來源 Chinese (Taiwan) (已轉換拼寫)

image/png

檔案歷史

檔案用途

全域檔案使用狀況

檔案來源 ^{Chinese (Taiwan) (已轉換拼寫)}