The short-time Fourier transform and the corresponding periodogram giv
e biased estimates of the instantaneous frequency (IF) if the IF in qu
estion is a nonlinear function of time. In the case of noisy signals,
the optimal choice of the window length, based on asymptotic formulae
for the variance and bias, can resolve the bias-variance trade-off usu
al for nonparametric estimation. However, the practical value of such
optimal estimator is not significant since the optimal window length d
epends on the unknown smoothness of the IF. The main goal of this pape
r is to develop an adaptive, periodogram-based IF estimator with a tim
e-varying and data-driven window length which is able to provide the a
ccuracy close to the one that could be achieved if the smoothness of t
he IF were known in advance. The developed algorithm uses only the est
imates of the IF and the formula for the Variance of these estimates.
Simulation shows good accuracy ability of the adaptive algorithm. (C)
1998 Elsevier Science B.V. All rights reserved.