Comparison of parametric and non parametric methods for signal time-frequency description

Abstract

This paper deals with comparing methods for time-frequency analysis such as time-frequency Fourier transform representation, multiple window method using Slepian sequences, time-frequency varying autoregressive process estimation and continuous wavelet analysis. To test their performance we apply these methods first on a simulated signal then on real recorded data from a metal plate acceleration by detonation products. For the autoregressive process we perform optimization of the lag order. The results are presented in graphical form and briefly discussed. Taking these results into account, combining the autoregressive process and wavelet transform is suggested.