Fall Detection System with Accelerometer and Threshold-based Algorithm

Most presently available fall detection systems that are marketed for commercial use predominantly consist of wearable technologies. These technologies often involve a device positioned on the wrist, which may lead to the occurrence of false positive alerts due to the movements of the wrist. This paper proposed a fall detection system that aims to improve both reliability and cost-effectiveness. The system is designed to promptly inform surrounding individuals of their need for assistance in emergency situations. The fall detection system we propose consists of an accelerometer and a gyroscope, which collectively calculate acceleration, orientation, and various other motion characteristics. The resulting system demonstrated a sensitivity of 90%, a specificity of 85%, and an accuracy of 87.5%.