Recursive identification ofwiener systems with general inputs

The paper concerns the identification of Wiener system consisting of a linear subsystem followed by a static nonlinearity. Based on integration by parts and Fourier transform, we give two recursive algorithms for estimating the coefficients of the linear subsystem with random inputs with possibly non-Gaussian distribution. The algorithm proposed in the paper includes the one given in Hu and Chen [2005] as a special case for the single-input single-output (SISO) systems, while for the multi-input multi-output (MIMO) systems it works in the case where there is no priori structural assumption made on nonlinearity. Two numerical examples are given to justify the theoretical analysis.