EMG Signal Analysis Using Sensor System and Edge Device for Wearable Applications

Abstract

Recently, bio-signal processing analysis using artificial intelligence is attracting great attention. In the case of bio-signal processing, it is a very difficult signal to analyze because it has a lot of noise induced by the users’ movement or other bio-signals. Therefore, different results depending on the person making the diagnosis may be obtained. To solve this problem, artificial intelligence through learning algorithm based on big data can be used. With this technique, it becomes possible to analyze bio-signals more simply and accurately without the help of a diagnostician. However, the artificial intelligence systems themselves are not suitable for wearable applications. In this paper, a sensor system was developed to be suitable for wearable applications in consideration of user convenience, and an edge device was also developed to analyze bio-signals with artificial intelligence. The electromyogram (EMG) signal was measured through the sensor system, and the analysis was performed with an edge device using field programmable gate array (FPGA). Through this, bio-signal analysis can be performed more easily, accurately, and at a user’s desired moment.