engleski

Detecting the number of components of multicomponent nonstationary signals using the Rényi entropy of their time-frequency distributions

In this paper, the Rényi entropy has been applied to different time-frequency distributions (TFDs) (namely, the Wigner-Ville distribution, the Choi-Williams distribution, and the spectrogram), with the purpose of determining which of the mentioned TFDs have best properties for estimation of the number of components when there is no a priori knowledge of the signal. The optimal parameter of the Rényi entropy, , for each TFD is determined. Simulations show that the Wigner-Ville distribution has best properties among the selected TFDs since it is invariant to time duration and frequency bandwidth variations of the analysed signal. The concept of a class, when the Rényi entropy is applied to TFDs, is also introduced.

Rényi entropy ; Time-Frequency Distributions ; Nonstationary signals

nije evidentirano