Extraction of frequency characteristics from EEG registnes using Wavelet transform

Abstract

The identification of oscillatory characteristics in (EEG) electroencephalographic signals has been widely studied due to the low cost of the technique and the use of information it provides. in this work, a scheme for the study of EEG registries using Wavelet in a group of healthy subjects - under two conditions: idleness and memory has been introduced. DWT Discrete Wavelet Transform was used in order to obtain coefficients with frequency and temporal information. Then, differences associated to condition change among cerebral rhythms in a set of analysis regions were extracted. Through the variance analysis, it was found that the largest contribution of variability is related to the r3 rhythm and it was verified that the behavior of a, rs and v is in agreement with prior research on memory processes. The proposed scheme constitutes the first step in the construction of an analysis methodology oriented to the study of subjects suffering from Alzheimer disease.