Recursive subspace identification of linear and nonlinear Wiener state-space models (Q5925917)

The paper deals with recursive identification of MIMO systems in the framework of subspace methods. The recursive formulations for the identification algorithms of the multivariable output-error state space class are derived using IV ideas and subspace tracking for the update of the SVD. A recursive algorithm for the identification of nonlinear Wiener type models with odd nonlinearity is also provided. Simulation examples illustrate the efficiency of the algorithms.