Multiple Windows for Estimation of Locally Stationary Transients in the Electroencephalogram

In this paper, multiple windows, optimal for locally stationary processes (MW-LSP) are used to estimate the spectrogram of the electroencephalogram (EEG) where we focus on the ability to estimate transient frequency changes. A peak of known frequency was evoked in the EEG spectrum in a predetermined time interval, by using a 9 Hz flickering light. We investigate the multiple windows corresponding to the mean squared error optimal time-frequency kernel for estimation of the Wigner-Ville spectrum. The kernel is optimal for a certain locally stationary process where the covariance function is determined by two one-dimensional Gaussian functions