Wearable low power pre-fall detection system with IoT and bluetooth capabilities

In today's society fall has become a very serious issue and also a threat to older generation. Fall leads to major injuries, physical disability and sometimes death. Therefore, there is a need to design a dependable embedded system device which will help in detecting the fall and further sends an emergency notification and thus helps in preventing the fall. Many researchers have considered human fall as an unpredictable danger to the life of older generation. Hence lot of research has been done in this area to develop a fall detection system. The existing fall detection systems are efficient but they lack in sensor optimization, early fall detection and efficient wireless communication. In this study, we have demonstrated a pre-fall detection system which detects human falls approximately 250 ms before it occurs. Early fall detection helps in preventing the subject from serious injuries. The designed system monitors user balanced and unbalanced state. Once the unbalance state is detected the system signifies it as a fall, thus gives milliseconds of time to trigger the safety devices like wearable airbag worn by a subject. On fall the system sends emergency notification to the care taker using either Internet of things (IoT) or Bluetooth low energy (BLE). The Hardware system is designed such that it consumes low power and it is a dependable embedded system with easy to wear capabilities for the subject. The designed system uses Arm Processor associated with motion sensors, communication sensors, Signal sensor and MicroSD Card. The software and hardware combination was developed to get optimal low power consumption by switching the CPU between active and sleep mode. The practical experiments performed on the designed system results in giving the 100% sensitivity and 98.07% specificity for fall detection. The wireless communication is efficiently designed such that the power is consumed only when the fall is triggered. The designed system acknowledges the difference between activity of daily living (ADL) like walking, sitting running, and climbing stairs with actual fall.