This monography is concerned with permutation entropy (PeEn) as a signal parameter for quantitative analysis of the electroencephalogram (EEG). The work focusses on two aspects: the efficient extraction of ordinal patterns from time series, as well as the interpretability of PeEn in the EEG context. The algorithms presented reduce the runtime of ordinal analysis methods significantly, and thus extend the maximum usable embedding dimension by an order of magnitude. In addition, the identification of characteristic regularities in the probability distributions of ordinal patterns in EEG allow to bridge the gap between PeEn and the Fourier transform.
«
This monography is concerned with permutation entropy (PeEn) as a signal parameter for quantitative analysis of the electroencephalogram (EEG). The work focusses on two aspects: the efficient extraction of ordinal patterns from time series, as well as the interpretability of PeEn in the EEG context. The algorithms presented reduce the runtime of ordinal analysis methods significantly, and thus extend the maximum usable embedding dimension by an order of magnitude. In addition, the identification...
»