Optimal kernels of time-frequency representations for signal classification
作者： M. Davy , C. Doncarli
关键词:
摘要: The use of distances between time-frequency representations (TFRs) has led to a formulation the problem non-stationary signals classification. We propose new method based upon optimization TFR, remaining in Cohen's group. show that radially Gaussian kernel and Fisher-like contrast criterion provide improved classification results.
sci-hub.se 参考文章(4)
H. Laurent, C. Doncarli, Stationarity index for abrupt changes detection in the time-frequency plane IEEE Signal Processing Letters. ,vol. 5, pp. 43- 45 ,(1998) , 10.1109/97.659547
F. Hlawatsch, G.F. Boudreaux-Bartels, Linear and quadratic time-frequency signal representations IEEE Signal Processing Magazine. ,vol. 9, pp. 21- 67 ,(1992) , 10.1109/79.127284
R.G. Baraniuk, D.L. Jones, A signal-dependent time-frequency representation: optimal kernel design IEEE Transactions on Signal Processing. ,vol. 41, pp. 1589- 1602 ,(1993) , 10.1109/78.212733
A.H. Costa, G.F. Boudreau-Bartels, Design of time-frequency representations using a multiform, tiltable exponential kernel IEEE Transactions on Signal Processing. ,vol. 43, pp. 2283- 2301 ,(1995) , 10.1109/78.469860