PERBANDINGAN METODE INDEPENDENT COMPONENT ANALYSIS DENGAN METODE PRINCIPAL COMPONENT ANALYSIS DALAM PROSES DENOISING SINYAL EEG
Total View This Week0
Total View This Week0
Institusion
Institut Teknologi Perusahaan Listrik Negara
Author
Muhammad, Givary
Yosrita, Efy
Aziza, Rosida Nur
Yosrita, Efy
Aziza, Rosida Nur
Subject
Teknik Informatika
Datestamp
2022-09-26 03:14:06
Abstract :
Electroencephalography (EEG) is an instrument that is used as brain activity recorder by showing the brain waves. EEG signal has a complex form, with small amplitude is accompanied by noise, and does not have any specific pattern, thus making it difficult to be analyzed visually. To increase accuracy and eliminate noise from EEG signals, a method is required to conduct a data cleaning process (denoising) such as Independent Component Analysis (ICA) or Principal Component Analysis (PCA). The accuracy of the result from the denoising process is not always satisfactory. This study will discuss ICA and PCA , and examine the performance of both methods in denoising by comparing their Signal-to-Noise Ratio (SNR) values. Data acquisition was conducted by utilizing a Brain Computer Interface (BCI) tools Emotiv Epoc+ 14 Channels. Implementation of this study will produce an application to simulate denoising process of an EEG data and shows comparison results from between both mehods. Testing results shows that Independent Component Analysis outmatch Principal Component Analysis with an average value of 92% from a number of tests.
Electroencephalography (EEG) is an instrument that is used as brain activity recorder by showing the brain waves. EEG signal has a complex form, with small amplitude is accompanied by noise, and does not have any specific pattern, thus making it difficult to be analyzed visually. To increase accuracy and eliminate noise from EEG signals, a method is required to conduct a data cleaning process (denoising) such as Independent Component Analysis (ICA) or Principal Component Analysis (PCA). The accuracy of the result from the denoising process is not always satisfactory. This study will discuss ICA and PCA , and examine the performance of both methods in denoising by comparing their Signal-to-Noise Ratio (SNR) values. Data acquisition was conducted by utilizing a Brain Computer Interface (BCI) tools Emotiv Epoc+ 14 Channels. Implementation of this study will produce an application to simulate denoising process of an EEG data and shows comparison results from between both mehods. Testing results shows that Independent Component Analysis outmatch Principal Component Analysis with an average value of 92% from a number of tests.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
