VGU RESEARCH REPOSITORY
Please use this identifier to cite or link to this item:
https://epub.vgu.edu.vn/handle/dlibvgu/905
Title: | Application of EEG signals in controlling IoT-based electronics devices | Authors: | Nguyen Xuan Tung | Keywords: | Brain-computer interface (BCI);Mind-controlled;Disabilities;Electroencephalograph (EEG);Control system;Microcontroller;Mental task | Issue Date: | 2017 | Publisher: | Vietnamese-German University | Abstract: | Brain-Computer Interface (BCI) is a system which can convert the human brain activities into control commands for other devices. For the last few decades, many researchers have put a great emphasis on this field of technology and developed useful applications, especially for paralyzed people who have severe disabilities, for instance mind-controlled wheelchair, mental typing system, etc. This master thesis proposes an electroencephalograph (EEG) based BCI system which assists disabled persons in urgent cases, specifically, help them to trigger a wifi-connected alarm bell or send an SOS message to their relatives via GSM network. In order to implement these to actions, users have to perform two imaginary movement tasks: “moving” their left or right arm. A low-cost wireless 11-electrodes EPOC headset by Emotiv company is used to acquire raw EEG. The signals are then sent into control signals. By applying Independent Components Analysis (ICA) technique, electrooculogram (EOG) artifacts contaminated in EEG data by eyes’ movement and blinking are eliminated to obtain cleaner signals. Power features of artifact-free signals are extracted and used as inputs of a Multi-Layer Perception (MLP) neural network, which plays the role of a classifier for the mental tasks. The intention of moving left and right arm are classified and translated into turning on the bell and sending sms message respectively. The proposed method requires obviously thinking practice of users. An individual set of EEG data patterns is also needed to train the MLP networks. Currently, experimental results showed the highest accuracy of 73.43% for classifying two categories of imaginary movements. |
URI(1): | http://epub.vgu.edu.vn/handle/dlibvgu/905 |
Appears in Collections: | Mechatronics & Sensor Systems Technology (MSST) |
Files in This Item:
File | Description | Size | Format | Existing users please Login |
---|---|---|---|---|
Application of EEG signals in controlling IoT-based electronics devices.pdf | 21.77 MB | Adobe PDF |
Page view(s)
93
checked on May 18, 2024
Download(s)
19
checked on May 18, 2024
Google ScholarTM
Check
Items are protected by © Copyright of Vietnamese - German University Library