SMART LIFE: A LIFESAVING WEARABLE SYSTEM FOR SENIOR CITIZEN
Keywords:IOT, Sensor Networks, LSTM, Apache, Fall Detection, MobiAct
Deterioration in an aged person's mobility, self-determination, and quality of lifestyle. This paper proposes a special Internet of Things-based system for recognizing indoor falls among the elderly by combining lightweight devices mobile sensing connections, big data, cloud computing, and smart appliances. For this, we use a wrist-worn sixLowPAN device equipped with an accelerometer with three axes to monitor the exact location and motion of senior citizens in real-time. A powerful IoT network analyzes the sensor signals is applying machine learning algorithms, which helps resulting in resulting in improved recognition of falls outcomes. For systems, we employ an incremental model with a long-memory framework. for the classification of falls, and economical Portable detecting gadget from Apache Flink and MbientLab, with a free software encoder. Using the initial data set, that is freely accessible "MobiAct," we evaluate the most effective Nyquist rate, sensor location, and multi-transmitting data modification. Our system for edge computing uses analytics on information streams in real-time to identify falls with a 95.87% efficiency ratio.