ARMA模型的Matlab代码

%--------------------------------------------------------------------------

% ARMA_LAST.M

% This function provides an ARMA spectral estimate which is maximum

% entropy satisfying correlation constraint (number of poles)and cepstrum

% constrains (number of ceros)

sigma,outn,outd

=arma_last(xinf,np,nq)

% The function requires:

% xinf(nsam) Input signal (real valued) of nsam samples

% nqOrder of the denominator (# of poles)

% npOrder of the numerator (# of ceros)

% The function provides:

% outd(nq)The numerator coefficients outn(1)=1

% outn(np)The denominator coefficients outd(1)=1

% sigmaThe square-root of input noise power

% The function requires spa_corc.m to compute the autocorrealtion matrix

% estimate

% Miguel Angel LAGUNAS and Petre STOICAJune 2007

%--------------------------------------------------------------------------

function

sigma,outn,outd

=arma_last(xinf,np,nq);

nsam=length(xinf);ra1=spa_corc(xinf,nq);nle=(nsam/2);

ra2=spa_corc(xinf,nle);=eig(ra2);

d1=diag(log(diag(d)));ca1=u*d1*u';clear u d1 d ra2 nsam;

for i=1:np

cuso(i)=0;

for j=1:nle-i;

cuso(i)=cuso(i)+ca1(j,i+j);

end;cuso(i)=cuso(i)/(nle);