Wearable health monitoring system based on human motion state recognition

The wearable health monitoring system is a typical application of wearable computing in the medical field. However, existing research often does not consider the characteristics of human physiological characteristics and exercise state in practical applications, and only judges the health of users from physiological data. The lack of information on the state of motion at that time caused a certain degree of misjudgment. For the coexistence of multiple types of devices and multiple transmission methods, this paper first proposes a wearable health monitoring system architecture based on human motion state recognition. Secondly, according to the characteristics of short-term persistence of daily activities of the human body, its motion state is divided into a steady state and an unstable state. The three-axis acceleration vector value is converted into the acceleration amplitude change amount, which eliminates the wearing correlation of the sensor coordinate system. Finally, we conducted a simulation experiment. The experimental results show that the algorithm achieves high accuracy in state recognition, and the recognition accuracy of running and walking is better than decision tree identification algorithm.