1
cspect Scilab Group Scilab Function cspect
3
cspect - spectral estimation (correlation method)
6
[sm,cwp]=cspect(nlags,ntp,wtype,x,y,wpar)
8
x : data if vector, amount of input data if scalar
10
y : data if vector, amount of input data if scalar
12
nlags : number of correlation lags (positive integer)
14
ntp : number of transform points (positive integer)
16
wtype : string : 're','tr','hm','hn','kr','ch' (window type)
18
wpar : optional window parameters for wtype='kr', wpar>0 and for
19
wtype='ch', 0 < wpar(1) < .5, wpar(2) > 0
21
sm : power spectral estimate in the interval [0,1]
23
cwp : calculated value of unspecified Chebyshev window parameter
26
Spectral estimation using the correlation method. Cross-spectral
27
estimate of x and y is calculated when both x and y are given.
28
Auto-spectral estimate of x is calculated if y is not given.
31
rand('normal');rand('seed',0);
33
//make low-pass filter with eqfir
34
nf=33;bedge=[0 .1;.125 .5];des=[1 0];wate=[1 1];
35
h=eqfir(nf,bedge,des,wate);
36
//filter white data to obtain colored data
37
h1=[h 0*ones(1:maxi(size(x))-1)];
38
x1=[x 0*ones(1:maxi(size(h))-1)];
39
hf=fft(h1,-1); xf=fft(x1,-1);yf=hf.*xf;y=real(fft(yf,1));
40
sm=cspect(100,200,'tr',y);
41
smsize=maxi(size(sm));fr=(1:smsize)/smsize;
50
Digital Signal Processing by Oppenheim and Schafer