Wavelet-Packet Identification of Dynamic Systems with Coloured Measurement Noise
This paper analyses the effect of coloured noise on a recently proposed technique for linear system identification in frequency subbands using wavelet packets. For this purpose, a simulation study involving the longitudinal dynamics of a flexible aircraft model is presented. The results reveal that the wavelet-packet identification outcome is robust with respect to changes in the spectral noise features. In particular, the identified frequency response is effectively smoothed in regions with poor signal-to-noise ratio. Finally, the results are favourably compared, in terms of resonance peak identification, with those obtained by standard time-domain identification methods.
KeywordsLeaf Node Wavelet Packet Decomposition Tree Frequency Subbands ARMAX Model
- 4.Coifman, R.R., Wickerhauser, M.V.: Entropy-based algorithms for best basis selection. IEEE Transactions on Information Theory 32, 712–718 (1992)Google Scholar
- 12.Ljung, L.: System Identification: Theory for the User, 2nd edn. Prentice-Hall, Englewood Cliffs (1999)Google Scholar