IDENTIFYING MIMO WIENER SYSTEMS USING SUBSPACE MODEL IDENTIFICATION METHODS

In this paper we show that the MOESP (Multivariable Output-Error State sPace) class of subspace model identification (SMI) schemes can be ex tended to identify Wiener systems, a series connection of a linear dyn amic system followed by a static nonlinearity. Methods are developed f or two cases: firstly when the nonlinearity is not an even function, a nd secondly for systems which include even nonlinearities. Simulations are used to demonstrate the efficacy of these algorithms, and to comp are them to existing, correlation based techniques.