Covariance Matching for Continuous-Time Errors-in-Variables Problems

Abstract

A covariance matching approach for continuous-time errors-in-variables problems is considered. The noise-free input signal is not explicitly modeled and is only assumed to be a stationary process. It is described how three different systems of equations in the model parameters are obtained and how the parameters are estimated based on these systems of equations. The approach is illustrated by some numerical studies.