The following information was submitted:
Transactions: INTERNATIONAL JOURNAL of CIRCUITS, SYSTEMS and SIGNAL PROCESSING
Transactions ID Number: 17-151
Full Name: Marco Schoen
Position: Professor
Age: ON
Sex: Male
Address: 921 S. 8th Avenue, Stop 8060
Country: UNITED STATES
Tel: 001 208 2824377
Tel prefix:
Fax: 001 208 2824538
E-mail address: schomarc@isu.edu
Other E-mails: sebaanis@isu.edu,kumaparm@isu.edu
Title of the Paper: Spatial filter masks optimization using genetic algorithm and modeling dynamic behavior of sEMG and finger force signals
Authors as they appear in the Paper: Anish Sebastian, Parmod Kumar, Marco P Schoen
Email addresses of all the authors: schomarc@isu.edu,sebaanis@isu.edu,kumaparm@isu.edu
Number of paper pages: 12
Abstract: Electromyography (EMG) signals are widely used for clinical and biomedical applications. One of the rapidly advancing fields of application of EMG is in the control of smart prosthetic devices for rehabilitation purposes. This paper presents the investigation of the use of System Identification (SI) for modeling sEMG-Finger force relation in the pursuit of improving the control of a smart prosthetic hand. Finger force and sEMG data are generated by having the subject perform a number of random motions of the ring finger to simulate various force levels. Post-processing of the sEMG signal is performed using spatial filtering. The linear and nonlinear spatial filters are compared based on the 'kurtosis' improvements and also based on the fit values of the models obtained using system identification, in particular the Hammerstein-Wiener models. The results of the modeling using linear spatial filters were found to be in the region of 30-45%, some of these linear spati!
al filter masks were selected randomly to investigate if there is any improvement in modeling the sEMG-force relation. The spatial filter masks are optimized using a Genetic Algorithm (GA) for two conditions; constrained and unconstrained. The model fit values of the identified models are used as the cost function in the GA optimization scheme. The results are compared to the reported filter mask values in the literature. The unconstrained GA based filter mask values and in some instances the constrained GA based mask values perform better than the filter masks reported in literature in 24 out of the 26 cases tested.
Keywords: Spatial Filtering, System Identification, Surface Electromyogram, Sensor Array, Genetic Algorithm, Hammerstein-Wiener Modeling
EXTENSION of the file: .pdf
Special (Invited) Session: Optimized spatial filter mask using genetic algorithm and system identification for modeling sEMG and finger force signals
Organizer of the Session: 658-232
How Did you learn about congress:
IP ADDRESS: 134.50.7.40
