Rankine - Signal Processing Research, Queensland University of Technology، Australia
Hassanpour - Signal Processing Research, Queensland University of Technology، Australia
Mesbah - Signal Processing Research, Queensland University of Technology، Australia
Boashash - Signal Processing Research, Queensland University of Technology، Australia

In this paper, we propose a novel method of modelling normal newborn EEG. The model is based on the results from the fractal dimension analysis of real EEG recordings from two newborns. The fractal dimension analysis was done using the Higuchi method for fractal dimension estimation. Comparison of signals and spectrums in the time domain, frequency domain and time-frequency domain between real and simulated signals show similarities validating the EEG model. This model can provide artefact free EEG signals for researchers in tuning parameters of an algorithm or evaluating automatic EEG detection algorithms such as seizure detection and epileptic spike detection.

EEG, Simulation, Fractal Dimension, Non- Linear, Non-Stationary