A step towards the development of VHDL model for ANN based EMG signal classifier

Abstract

The artificial neural network (ANN) is an information processing model which is developed from the inspiration of practical biological nervous system. ANNs are analogous to the human brain, which can perform a variety of complex tasks if configured properly through a learning process. The research work involves with the utilization of ANN as a classifier for hand motion detection by using Electromyography (EMG) signals. A feed-forward ANN with back-propagation learning algorithm is used for the classification of EMG signals. This paper illustrates the modeling of the neural network based classifier using Hardware Description Language (HDL) for hardware realization. VHDL (Very High Speed Integrated Circuit Hardware Description Language) has been used to model the algorithm and which can be implemented into the target device FPGA (Field Programmable Gate Array). The development process and simulation output are presented in details with the architectural design of the neural network. The designed model has been synthesized and fitted into Altera's Stratix III, chipset EP3SE50F780I4L using the electronic design automation (EDA) software Quartus II version 9.1 Web Edition.