Recursive Parameter Identification of Wiener Systems with Discontinuous Piecewise-linear Memoryless Block and Observation Noise

The paper deals with parameter identification of the Wiener system with discontinuous piecewise-linear memoryless block and with observation noise. An upper bound for the discontinuous point d⁺ (and d^-) of the memoryless block is recursively estimated, and the estimate is proved to converge to a magnitude greater than d⁺(d^-). Recursive estimates are given for the coefficients of the linear subsystem and for the six unknown parameters contained in the discontinuous piecewise-linear memoryless block. All parameter estimates are proved to be strongly consistent. A numerical example is given, and the computational results are consistent with the theoretical analysis.