Time-varying bispectral analysis of non-stationary signals

Abstract

In this paper, we combine the Higher Order Statistics (HOS) and evolutionary spectral analysis approaches for non-stationary signals. HOS are useful for the analysis of non-Gaussian, non-linear and non-stationary properties of processes. Time-frequency analysis is used to reveal the time-varying features of a signal. We present a method for the calculation of a Time-Dependent higher order spectrum Evolutionary Bispectrum - based on the multi-window Gabor expansion.