Dissertation > Industrial Technology > Radio electronics, telecommunications technology > Communicate > Communication theory > Signal processing

Analysis and Research on EEG Signals Based on HHT Algorithm

Author FanYingYing
Tutor LiuZhuoFu
School Harbin University of Science and Technology
Course Communication and Information System
Keywords Hilbert-Huang Transform Electroencephalogram empirical modedecomposition intrinsic mode function noise suppression
CLC TN911.7
Type Master's thesis
Year 2012
Downloads 42
Quotes 0
Download Dissertation

Electroencephalogram(short for EEG) contains a lot of physiology andpathology information. The information obtained can not only give us theevidence of clinical diagnosis, but also provide an effective measure of adjuvanttherapy of certain brain diseases for the doctors. But if we want to get the usefulinformation, we have to process the EEG signals first. The traditional method ofprocessing signals such as the Fourier Transform, are usually suit to the linearand steady signals. But the EEG signals are non-linear and not steady, FourierTransform is not useful for the EEG signals. Hilbert-Huang Transform(short forHHT) has the ability of self-adaption and suit to all kinds of signals, so we adoptit to process the EEG signals in this paper.The main contents in this paper can be summarized as follows:First, we introduce the basic knowledge of the EEG signals and the researchon the EEG signals at home and abroad. Then we describe the process and thecharacteristic of empirical mode decomposition and compare wavelet transform,the Fourier Transform and HHT. By comparison, we get that HHT is more suit toprocess the EEG signals. Then we put forward a new method of removing noise.This method is combined empirical mode decomposition and Monte Carlo, thecore of which is to set up a white noise base. Then compare the signal with thewhite noise base to remove noise. In order to validate whether the method iseffective, we do the simulation under the MATLAB platform. After that, doHilbert-Huang Transform and Fourier Transform with the EEG signals whichhave been removed noise, and the Hilbert spectrum and the Fourier spectrum willbe gained, respectively. Finally, we analyze the EEG signals through the Hilbertspectrum and the Fourier spectrum and obtain the useful information. Throughthe simulation, we prove that Hilbert-Huang Transform is useful for processingthe EEG signals, also prove that it is more excellent than Fourier Transform. We combine MATLAB and VC++platform and realize the processing of the EEGsignals without MATLAB environment to improve the speed of processingsignals in this paper.

Related Dissertations
More Dissertations