We developed a low power kinematic sensor, ActimedARM, incorporating three-axis accelerometer and magnetometer, a microcontroller ARM3, a ZigBee wireless communication and μSD memory storage. With embedded algorithms it can detect in real time the postures of the subject. A preliminary assessment conducted on 12 subjects reached a 97% correct classification rate. The device exhibits 32 days of autonomy on a 3600 mAh capacity battery, which makes it convenient for field experiments in true daily life.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.1109/EMBC.2013.6609884 | DOI Listing |
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!