Time-varying system identification and model validation using wavelets

  title={Time-varying system identification and model validation using wavelets},
  author={Michail K. Tsatsanis and Georgios B. Giannakis},
  journal={IEEE Trans. Signal Processing},
Parametric identification of time-varying (TV) systems is possible if each TV coefficient can be expanded onto a finite set of basis sequences. The problem then becomes time invariant with respect to the parameters of the expansion. In this paper, we address the question of selecting this set of basis sequences. We advocate the use of a wavelet basis because of its flexibility in capturing the signal’s characteristics at differerent scales, and discuss how to choose the optimal wavelet basis… CONTINUE READING
Highly Cited
This paper has 190 citations. REVIEW CITATIONS
76 Citations
20 References
Similar Papers


Publications citing this paper.
Showing 1-10 of 76 extracted citations

190 Citations

Citations per Year
Semantic Scholar estimates that this publication has 190 citations based on the available data.

See our FAQ for additional information.


Publications referenced by this paper.
Showing 1-10 of 20 references

An ordertesting criterion for mixed autoregressive moving average processes

  • C. J. Hams, P. E. Wellstead
  • Int . J . Conrr .
  • 1992

Detection of a transient signal of unknown scaling and arrival time using the discrete wavelet transform,

  • I l l M. Frisch, H. Meser
  • Proc. ICASSP,
  • 1991

Time - dependent ARMA modeling of nonstationary signals

  • B. Quinn
  • Acoust . , Speech , Signal Processing
  • 1991

Orthogonal bases of compactly supported wavelets,

  • I. Daubechies
  • Commun. Pure Appl. Math., vol. 41,
  • 1988
2 Excerpts


  • E. W. Kamen, P. P. Khargonekar, R K.
  • “A transfer function approach to linear time…
  • 1985

Similar Papers

Loading similar papers…