Abstract:
In this paper,a parameter estimation method for continuous Wiener model containing random noise is proposed.The method uses M periodic sequences of different magnitude as input signals.A linear submodel is sepa-rated from the original after analysis.By using a frequency estimation method,the parameters for the linearsubmodel are obtained.The coefficients of nonlinear characteristics are estimated further by using least squaresmethod.The relationship between the noise signal after transformation and the original noise signal is discussed andcorrespondingly two lemmas are given and proved in this paper.The simulation results show that both the accura-cy and computing speed are rather high with the method.