engleski

Different Analysis Procedures In EEG Data Signal Processing

In the electroencephalography (EEG) data signal processing [1], commonly used methods utilize time and frequency domain analysis. Analysis in time domain gives an amplitude value of the signal in each time moment. As opposed to that, analysis in frequency domain indicates which frequencies are present in the signal. Furthermore, lack of frequency-information in the time domain analysis and lack of time-information in frequency domain analysis in nonstationary signals where frequency content varies with time leads to a usage of joint time-frequency representation (TFR)[2]. Such representation, which is localized in both time and frequency, is obtained using a distribution whose domain is the two-dimensional (t, f) space called time-frequency distribution (TFD). Apart of the most commonly used TFD - Wigner-Ville distribution (WVD), in this paper EEG signal is presented and compared with several different TFDs. Additionally, some differences between different types of analyses, such as empirical mode decomposition, wavelet etc., are explained.

Time domain, frequency domain, time-frequency domain

nije evidentirano