UOCAL STATIONARITY OF L2(R) PROCESSES
Francisco Moreno, Isabel M. G. Lourtie, Jorge Buescu · 2002
This paper shows how the sampling theorem relates with the variations along time of the second order statistics of L 2 (R) non stationary processes. As a consequeRl; �, and minly ?�e toe positive semidefiniteness of autocorrelati? R funCtions, . It IS POSI ble to conclude if a nonstationary process IS locally staboRluy (u., if its second order statistics vary slowly along time) by the direct observation of its 2-dimension power spectrum or its Wigner dis tribution. A simple example iIIustRltes how two different strategies for the estimation of autocorre lation functions from a small num ber of data can lead to opposite results in terms oflocal stationarity.