M.S. (Master of Science)
Department of Electrical Engineering
This research will cover Electromyography (EMG) and its use in prosthetics. There are many upper limb amputees that use EMG to control a prosthetic limb. While many of these prostheses are proportionally controlled, meaning they can control one movement proportional to the EMG amplitude, there is a continual need to improve function for quality of life. Proportional control is traditionally utilized for opening and closing of the hand. It is the desire of engineers to find ways to add more functionality without sacrificing reliability. There are systems that do these additional actions, but they have difficulty with reliability because of using a fixed, predefined, set of features. Through using wavelet transforms and an eigen-analysis of the feature space, this research explores the use of a method to develop a customized feature set. By selecting components of a signal that lend themselves to an optimum energy distribution of classes, a custom feature can be extracted. This extracted feature is based on the users data, allowing customization of the prosthetic control and improving trust with the device.
Keywords: Electromyography (EMG), Wavelet Transform (WT), Short Time Fourier Transform (STFT), Prosthetics, Eigenvectors, Eigenvalues, Statistical Detection Theory, Neyman-Pearson, Bayesian Classification
Bonnen, Luke S., "Wavelet-Eigen Analysis for Customizable Flexion Classifier Based on Electromyography Signals" (2019). Graduate Research Theses & Dissertations. 6877.
Northern Illinois University
Rights Statement 2
NIU theses are protected by copyright. They may be viewed from Huskie Commons for any purpose, but reproduction or distribution in any format is prohibited without the written permission of the authors.