Dissertation/Thesis Abstract

P300 Control Matrix: A Novel Approach to P300 Speller Matrix
by Odelade, Mobolaji, M.S., North Carolina Agricultural and Technical State University, 2018, 58; 10976563
Abstract (Summary)

Over the years, researchers have been able to prove Brain Computer Interface (BCI) -P300 Speller as an effective communication tool. The first P300 speller was developed by Farwell and Donchin (1988), using the oddball paradigm to evoke a P300 response from a speller matrix. This P300 speller matrix has been a strong basis for studies that aimed at using BCI-P300 protocol for spelling, cursor movement, internet navigation or even control and manipulation of devices. However, application of P300 based BCI to controlling and manipulation of devices often involves the user relating with multiple interfaces. These multiple interfaces could be a distraction or have negative effects on the user (Fazel-Rezai et al. 2012) and as a consequence hinders the evoking of P300 potential and causing inaccurate classification. For this research, a novel P300 control matrix is developed by replacing the alphabets in the traditional P300 speller matrix with arrow images. Then the novel P300 control matrix was investigated to compare the P300 latency and amplitude to that of the traditional P300 speller matrix. The elements in the novel P300 control matrix were in form of arrows facing upward, left, right and downward directions, while elements in the P300 speller matrix were alphabets U, L, R and D for the upward, left, right and downward directions respectively. The participants were presented with a set of randomly sequenced directions, and each participant decides which of the arrows or letters to focus on based on the direction presented to them. Electroencephalography (EEG) was used to record the brainwaves using the international 10-20 system of electrode placement. This research is potentially a more efficient approach for controlling devices using P300-based BCI systems by eradicating the need for multiple interfaces associated with BCI-robotic control systems that are based on P300 speller.

Indexing (document details)
Advisor: Seong, Younho
Commitee: Jiang, Steven, Yi, Sun
School: North Carolina Agricultural and Technical State University
Department: Industrial and Systems Engineering
School Location: United States -- North Carolina
Source: MAI 58/03M(E), Masters Abstracts International
Subjects: Computer Engineering, Robotics
Keywords: Bci, Eeg, Robotic control
Publication Number: 10976563
ISBN: 978-0-438-73897-3
Copyright © 2020 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy