Activity recognition using an e-textile data acquisition system

Abstract

In our research we have proposed implementation of an activity recognition system based on a data acquisition system with e-textile components. One of the most important requirement of data acquisition systems used in human activity recognition is their level of acceptability or their level of obstruction [2, 4]. Starting from this important requirement, we designed a wearable data acquisition system using e-textile components, in the form of buttons with an attractive aesthetic, easy to integrate in clothing objects of a person. In our experiments the system was built on a cuff which can be easily attached to the hand or foot. Our system contains, on the wearable part, three Lilypad type, e-textile sensors (acceleration on three-axis, light, and temperature), a Lilypad Arduino microcontroller module, and a communication module which consist in a LilyPad XBee module with an XBee 1mW Trace Antenna module integrated. For receiving data, on the fixed part of the system, we have an Arduino communication module with an XBee module attached, and an Arduino Uno platform connected to a PC through a serial port. For activity recognition, data files are transferred to a PC, and data is processed in the Matlab using artificial neural networks as classifier.