Please use this identifier to cite or link to this item: http://dspace.unimap.edu.my:80/xmlui/handle/123456789/20587
Title: Synchronous brain machine interface design using focused time delay networks
Authors: Hema, Chengalvarayan Radhakrishnamurthy
Paulraj, Murugesa Pandiyan, Assoc. Prof. Dr.
Sazali, Yaacob, Prof. Dr.
Abdul Hamid, Adom, Assoc. Prof. Dr.
Ramachandran, Nagarajan, Prof. Dr.
hema@unimap.edu.my
paul@unimap.edu.my
Keywords: Brain Machine Interface (BMI)
Neural network
Wheelchair
Focused time-delay neural network (FTDNN)
Issue Date: 27-Feb-2012
Publisher: Universiti Malaysia Perlis (UniMAP)
Series/Report no.: Proceedings of the International Conference on Man-Machine Systems (ICoMMS 2012)
Abstract: Focused time delay neural network based design for a four-state Brain Machine Interface (BMI) to drive a wheelchair is analyzed. Motor imagery signals recorded noninvasively using two bipolar electrodes are used in the study. The performance of the proposed dynamic classifier is compared with a static feed forward neural classifier. Data collected from 10 subjects is used in this study. Average classification performance in the range of 93% to 100% is achievable. Experiment results show that the focused time delay neural network model is suitable for a four-state BMI design.
Description: International Conference on Man Machine Systems (ICoMMS 2012) organized by School of Mechatronic Engineering, co-organized by The Institute of Engineer, Malaysia (IEM) and Society of Engineering Education Malaysia, 27th - 28th February 2012 at Bayview Beach Resort, Penang, Malaysia.
URI: http://dspace.unimap.edu.my/123456789/20587
Appears in Collections:Conference Papers
Abdul Hamid Adom, Prof. Dr.
Sazali Yaacob, Prof. Dr.
Paulraj Murugesa Pandiyan, Assoc. Prof. Dr.

Files in This Item:
File Description SizeFormat 
132-63582_Synchronous Brain Machine Interface Design using Focused Time Delay Networks.pdfAccess is limited to UniMAP community45.26 kBAdobe PDFView/Open


Items in UniMAP Library Digital Repository are protected by copyright, with all rights reserved, unless otherwise indicated.