Low power consumption fall detection using three features
Maria Seraphina Astriani, Raymond Bahana, Andreas Kurniawan, Hyun-Kyo Lee · IOP Conference Series Earth and Environmental Science · 2021
Abstract The increasing of longevity rates contributed to elderly and in line with the increased of medical needs especially on health care and monitoring systems area. To facilitate continuous monitoring and to address the need of healthcare which are non-invasive, affordable, easy-to-use, and non-invasive healthcare solutions are becoming increasingly important. Smartphone is a perfect device to detect human fall because it has various sensors (accelerometer, gyroscope, GPS, and many more), already been accepted by most of the people, and can reduce the electronic waste by giving a smartphone a second chance to become a health care and monitoring system. The research proposed a low power consumption human fall detection by using three features: Signal Vector Magnitude (with modification), Alim, and Tilt Angle Change as a solution to overcome two problems (health care and environment) by using the device that already been existed to reduce the electronic waste. Our proposed solution was able to reach 0.97 of accuracy result and work on the smartphone (low power device) for green computing.