cspect

cspect — spectral estimation (correlation method)

Calling sequence

[sm,cwp]=cspect(nlags,ntp,wtype,x,y,wpar)  

Parameters

x : data if vector, amount of input data if scalar
y : data if vector, amount of input data if scalar
nlags : number of correlation lags (positive integer)
ntp : number of transform points (positive integer)
wtype : string : 're','tr','hm','hn','kr','ch' (window type)
wpar : optional window parameters for wtype='kr', wpar>0 and for wtype='ch', 0 < wpar(1) < .5, wpar(2) > 0
sm : power spectral estimate in the interval [0,1]
cwp : calculated value of unspecified Chebyshev window parameter

Description

Spectral estimation using the correlation method. Cross-spectral estimate of x and y is calculated when both x and y are given. Auto-spectral estimate of x is calculated if y is not given.

Examples



rand('normal');rand('seed',0);
x=rand(1:1024-33+1);
//make low-pass filter with eqfir
nf=33;bedge=[0 .1;.125 .5];des=[1 0];wate=[1 1];
h=eqfir(nf,bedge,des,wate);
//filter white data to obtain colored data 
h1=[h 0*ones(1:maxi(size(x))-1)];
x1=[x 0*ones(1:maxi(size(h))-1)];
hf=fft(h1,-1);   xf=fft(x1,-1);yf=hf.*xf;y=real(fft(yf,1));
sm=cspect(100,200,'tr',y);
smsize=maxi(size(sm));fr=(1:smsize)/smsize;
plot(fr,log(sm))
 
  

See also

pspect

Author

C. Bunks ; Digital Signal Processing by Oppenheim and Schafer