Design And Implementation Of Human Movement Classifier In Monitoring System

Rapid technological advances have prompted the development of a wide range of telemonitoring systems. The home telecare paradigm is a promising approach for reducing the reliance on institutionalized health care costs by avoiding unnecessary hospitalization. With the aid of home-based health monitoring network, it is easier to detect the ability of elderly people to live independently and provide an automated system of supervising functional status over extended time periods. However, this application brings new challenges to the research on ambulatory monitoring, that is, how to acquire physiological data from a subject who is free to go about his normal activities. This thesis focuses on this issue by the aid of wearable sensor.First of all, to ensure the wearer's activity and level of comfort, this monitoring system is associated with the data acquired from a single triaxial accelerometer (TA) unit worn at the waist. The enhancements in microelectromechanical systems (MEMS) technology have made possible the manufacture of miniaturized, low cost accelerometers. These instruments also demonstrate a high degree of reliability in measurement and capacity of offering a number of desirable features in monitoring of human movement. Wireless communication in our system is facilitated by the IEEE 802.15.4 standard to provide a reliable transportation of system results between the TA device and the external base stations.Considering the factor of run-time efficiency, the real-time data processing is divided into two parts, the preprocessing module and the movement classifier module. And the core is the design of classifier. Since in previous study, data processing and analysis are performed offline, making such devices impractical for continual monitoring, our system aims at performing the signal processing onboard the wearable unit using embedded intelligence. In this way, the system distinguishes between periods of activity and rest, recognizes the postural orientation of the wearer. And the system has been designed to identify activities of standing, sitting, lying, transition and detect events such as walking and falls.