Recursive subspace identification of linear and non-linear Wiener state-space models

The problem of MIMO recursive identiication is considered and analyzed within the framework of subspace-based state space system identiication (4SID). With respect to previous contributions to this area, in this paper the use of recent signal processing algorithms for the recursive update of the SVD is proposed. To accommodate for arbitrary correlation of the disturbances, an instrumental variable (IV) approach is followed. In particular, three diierent recursive formulations for the subspace identiication algorithms of the MOESP class are given. As a byproduct, a recursive algorithm for the identiication of nonlinear models of the Wiener type is also obtained.