Analysis of the asymptotic properties of the MOESP type of subspace algorithms

Analysis of the asymptotic properties of the MOESP type of subspace algorithms Abstract The MOESP type of subspace algorithms are used for the identiication of linear, discrete time, nite dimensional state space systems. They are based on the geometric structure of covariance matrices and exploit the properties of the state vector extensively. In this paper the asymptotic properties of the algorithms are examined. The main results include consistency and asymptotic normality for the estimates of the system matrices, under suitable assumptions on the noise sequence, the input process and the underlying true system.