System identification using a wavelet-based approach

In this paper, an approach to system identification using the wavelet transform is outlined. The method is based on system identification algorithms that belong to the class of modulating function methods. In particular, the Pearson-Lee (1985) method is modified to make use of wavelet basis functions which provide a more precise time window to process the plant input-output data. It is proposed here that replacing these modulating functions with wavelet basis functions will provide data filtering which is less sensitive to noise and initial conditions