System identification with denoising
Document Type
Conference Proceeding
Publication Date
1-1-2000
Abstract
When the signal-To-noise ratio (SNR) is low, classical system identification methods can not produce accurate results. The results can be improved by using denoising methods with time-frequency decompositions. The chirp signal is used as a training sequence to make the time-frequency domain denoising possible. Chirplet decomposition is proposed for separation of signal and noise components. The results are compared with the Gabor transform denoising. The chirplet denoising method proposed here is less sensitive to SNR changes than the Gabor denoising proposed before. Also, the accuracy of the estimates in chirplet case is superior to the Gabor transform method.
Identifier
0033693084 (Scopus)
ISBN
[0780362934]
Publication Title
ICASSP IEEE International Conference on Acoustics Speech and Signal Processing Proceedings
External Full Text Location
https://doi.org/10.1109/ICASSP.2000.862047
ISSN
15206149
First Page
576
Last Page
579
Volume
1
Recommended Citation
Bultan, Aykut and Haddad, Richard A., "System identification with denoising" (2000). Faculty Publications. 15724.
https://digitalcommons.njit.edu/fac_pubs/15724
