Ping He (Advisor)
Master of Science in Engineering (MSEgr)
The Primary Objective of this research is to implement an automatic method for selecting the most optimal EEG channels for task classification purposes. The secondary objective of this research is to choose the most optimal EEG rhythm from which the optimal EEG channels would be selected automatically. The automatic selection of the optimal channels is enabled by implementing the Common Spatial Patterns algorithm (CSP). Common spatial analysis is performed on the data recorded. By choosing the channels with high spatial pattern values the optimal channels are chosen. The optimal frequency bands are chosen by splitting the data from a single channel into different frequency bands such as the alpha, beta, theta and gamma bands and classifying the data obtained from each bands. The feature vector for a particular task is computed by application of the common spatial filter on the data recorded. A linear Fisher's discriminant method is used for classification process. The entire data analysis for this project is done using MATLAB.
Department or Program
Department of Biomedical, Industrial & Human Factors Engineering
Year Degree Awarded
Copyright 2007, all rights reserved. This open access ETD is published by Wright State University and OhioLINK.