Identification of Wiener and Hammerstein Systems with Rate Saturation

This paper proposes the use of the coherence function for distinguishing between a Wiener structure and a Hammerstein structure for systems with rate saturation nonlinearity. This is an important extension in the application of the coherence function from the usual case of static nonlinearity to cover dynamic nonlinearity. The test capitalizes on the same set of data generated when estimating the best linear approximation of the system. The identification of the system parameters through the best linear approximation is also analyzed in terms of the sensitivity and convergence properties. The effectiveness of the approach is illustrated through simulations. The results show that the proposed approach combining the structure identification and parameter estimation outperforms an existing method.