Fall Detection Algorithm based on Peaks of Voltage Measurements from the Accelerometer

According to the Korea Census Bureau, the number of elderly people above the age 65 will reach more than 10 million in 2026. It means that more than 21% of the population in KOREA will reach over 65 years old. It becomes extremely important to take care of the elderly people for the case of health emergency; such as fall or heart attack. In this paper, a simple fall detection system that detects an actual fall direction is implemented. In order to detect the fall situation, 3-axis acceleration sensor (MMA7331) is used in the system. The fall detection algorithm that can classify fall directions such as front, back, left and right fall is proposed. The direction of the fall is decided by examining the acceleration peaks of X and Y directions of the sensor. It is shown that the proposed algorithm successfully detects the front and back fall direction with probability of 96% and 98%, respectively.