Time delay estimation with higher order moments in epilepsy [EEG analysis]

Author

Kinie, A. ; Carrault, G. ; Wendling, F.

Author_Institution

Lab. Traitement du Signal et de l´´Image, Rennes I Univ., France

fYear

1994

Firstpage

233

Abstract

Two approaches based on higher order moments (bispectrum) for time delay estimation are compared: a conventional approach which uses the fast Fourier transform (FFT) and a parametric method, The results obtained from a same simulation protocol (nonlinear transmission channels, different values for signal to noise ratio, etc...) on simulated signals point out the contribution as well as the limits of each method. Some results on real signals complete this preliminary study

Keywords

electroencephalography; fast Fourier transforms; medical signal processing; EEG analysis; bispectrum; electrodiagnostics; epilepsy; fast Fourier transform; higher order moments; nonlinear transmission channels; parametric method; signal to noise ratio; simulated signals; simulation protocol; time delay estimation; Brain modeling; Delay effects; Delay estimation; Electroencephalography; Epilepsy; Fast Fourier transforms; Protocols; Random processes; Signal processing; Signal to noise ratio; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

Engineering in Medicine and Biology Society, 1994. Engineering Advances: New Opportunities for Biomedical Engineers. Proceedings of the 16th Annual International Conference of the IEEE

Print_ISBN

0-7803-2050-6

Type

conf

DOI

10.1109/IEMBS.1994.412067

Filename

412067