Adaptive system identification based on generalized wavelet decomposition (Q1892056)

An adaptive structure based on a wavelet packets filter bank is proposed. The system identification of a finite impulse response is investigated by utilizing the presented adaptive scheme. In order to attain the stabilized convergence of the adaptive filters, a regularization parameter is introduced for each wavelet packet to minimize the mean squares error of the adaptive filter weights. The effectiveness of the proposed approach is demonstrated through examples.