G.V. Tcheslavski and A.A. (Louis) Beex (USA)
Phase synchrony, EEG, analysis length, bandwidth.
The phase synchrony analysis of stochastic time series is
considered with a view towards its application in EEG
(electroencephalogram) processing. A Phase Synchrony
Processor is proposed and its properties are examined on
the basis of known signals. The observed dependence of
the phase synchrony coefficient on the analysis
parameters, such as the filter bandwidth and the length of
the time-window, may - especially in low synchrony cases
- lead to biased results. The analysis parameters can be
chosen judiciously based on the results of a phase
synchrony study of artificially generated signals with
known phase synchrony. To illustrate the importance of
the parameter choices available, a phase synchrony
coefficient analysis is presented based on actual EEG.