DEVELOPMENT OF OPEN HARDWARE-BASED WIRELESS ELECTROMYOGRAPHY (EMG) SYSTEM FOR BIO-FEEDBACK DURING MUSCLE ACTIVITY QUANTIFICATION

• Main Author: Amandi R, Rachmat
• Format: Final Project
• Language: Indonesia
• Subjects: Teknik (Rekayasa, enjinering dan kegiatan berkaitan)
• Online Access: https://digilib.itb.ac.id/gdl/view/47966
• Institution: Institut Teknologi Bandung

In general, EMG signal frequencies are in the range of 5 Hz to 128 Hz. In order to properly measure EMG signals, the sampling frequency of used EMG device must be at least 256 Hz. Recently most of the EMG devices could sample EMG signals up to 2048 Hz. Higher sampling frequency of EMG device results in sharper measured EMG signal. However, the computational load required to process the EMG signal will be higher. Currently EMG signals is used for many purposes, including monitoring muscle activities during medical rehabilitation and as control signals for gaming and robotics. The increasing usage of EMG signals has led to the development of open hardware EMG devices. One of the open hardware EMG device with sampling frequency that could reach 256 Hz is OpenBCI Cyton. OpenBCI Cyton main component is a digital bioamplifier that functions as amplifier and ADC with standard sampling frequency of 250 Hz. OpenBCI Cyton sampling frequency can reach up to 1 kHz while connected to OpenBCI WiFi Shield. The main purpose of this research is to design open hardware-based wireless EMG device. Sampling frequency test for OpenBCI WiFi Shield resulting in a maximum effective sampling frequency of 500 Hz. OpenBCI WiFi Shield configuration resulting in higher background noise amplitude than the standard configuration, which are 11,811 mV and 4,088 mV respectively. Power spectrum analysis shows that the standard configuration resulting in sharper decline of signal power above 60 Hz. The use of OpenBCI WiFi Shield successfully reduce the aliasing effect due to undersampling, with the disadvantage of higher background noise amplitude. Statistical analysis shows that differences in sampling frequency resulting in a significant difference on the EMG signal.