Asymptotic accuracy of covariance function based errors-in-variables system parameter estimates

A covariance matching approach for estimating the parameters in errors-in-variables systems is analyzed and the asymptotic accuracy of the parameter estimates is derived. The computation of the asymptotic covariance matrix of the parameter estimates is given in an algorithm and it is shown that the approach is asymptotically statistically efficient in a certain sense. An algorithm for computing the covariance matrix of the input-output covariances is also given.