www.pudn.com > Gngauss.rar > Gaus_mar.m, change:2003-04-13,size:481b

function [X]=gaus_mar(X0,rho,N) 
% [X]=gaus_mar(X0,rho,N) 
%       GAUS_MAR  generates a Gauss-Markov process of length N. 
%           The noise process is taken to be white Gaussian 
%           noise with zero mean and unit variance. 
for i=1:2:N, 
  [Ws(i) Ws(i+1)]=gngauss;        % Generate the noise process. 
X(1)=rho*X0+Ws(1);                % first element in the Gauss--Markov process 
for i=2:N, 
  X(i)=rho*X(i-1)+Ws(i);          % the remaining elements