Heterogeneous Sensor Data Fusion for Human Falling Detection

With the continuous improvement of human living standards, population aging has become a global development trend. At present, China entered an society, health and safety elderly have focus social concern. Due to physiological structure decline physical function, probability frequency accidental falls in are very high. Under above background, purpose this study is based on heterogeneous sensor data fusion algorithm intelligent wearable network. This article proposes wireless body area network technology, constructs high-precision stable activities daily (ADLs) fall monitoring system. We first select three-axis acceleration sensor, magnetic angular velocity monitor elderly. Then, we use Bluetooth transmit collected by sensors smartphones, communicate with service centers users through mobile phone communication network, Family members interact form city technology. Our proposed approach Kalman filter algorithm, which can reduce system noise improve stability The experimental results demonstrate that detection implemented well detect elderly, sensitivity specificity 98.7% 98.5% respectively. high research value practical application significance protecting healthy life