POWER SPECTRAL DENSITY (PSD) OF ELECTROENCEPHALOGRAPH (EEG) APPLICATION FOR HAND MOVEMENTS ON BRAIN-COMPUTER INTERFACE (BCI)

There are number of cases which causes human is not able to use some or all part of their body, therefore, moving some devices with mind order can be one of the solutions. Brain-Computer Interface is a technology that can make this happen. Brain-Computer Interface (BCI) connects commands and respons...

Full description

Saved in:
Bibliographic Details
Main Author: Dwiratna, Pranacintya
Format: Final Project
Language:Indonesia
Online Access:https://digilib.itb.ac.id/gdl/view/53689
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Institut Teknologi Bandung
Language: Indonesia
Description
Summary:There are number of cases which causes human is not able to use some or all part of their body, therefore, moving some devices with mind order can be one of the solutions. Brain-Computer Interface is a technology that can make this happen. Brain-Computer Interface (BCI) connects commands and responses that given by brain waves that can be measured using Electroencephalograph (EEG) and turn it into information that machine could understand. The purpose of this research is to understand how to read PSD information from EEG data and to obtain an accurate method of processing and classification of EEG data to be used in BCI to distinguish brain waves of right and left hand motor imagery. The EEG data used in this research is a set data from a previous research by Yuriy Mishchenko in 2018. There are several stages that are used in data processing. The stages are pre-processing with Fast Fourier Transform (FFT), feature extraction with Power Spectral Density (PSD), and classification with Support Vector Machine (SVM) using MATLAB and Python. The peak frequency of the PSD chart used as a feature in the SVM classification can get an accuracy of up to 59.23%.