Time-frequency dependent transformations and their application in digital signal processing

Slobodan Simić

Bojan Zrnić

This paper presents an approach to the non-stationary signal and process analysis based on time-frequency dependent transformations. The focus is placed on the Wigner distribution, its time and frequency smoothed versions and the short-time Fourier transform the results of comparison of the aforementioned distributions concerning the resolution, interferences, pushing of the side lobes and characteristics are also presented.

Ključne reči:   time-frequency distribution, Wigner distribution, pseudo-Wigner distribution, short-time Fourier transform, resolution, interference, windo-wing function.

FUL TEXT