| The increasing population of elderly people needs help to support their health, independency and safety. The monitoring of human movement can provide valuable information about individual activities and maybe further used for tele-care. This thesis describes the implementation of a system for motion monitoring (fall detection, walking, running, sleeping etc.) in real-time using wireless sensor network technologies. The system uses motes (from Crossbow) to collect motion data. The data is delivered to a desktop computer and is analyzed in real-time. Some standard techniques such as FFT, median filter, low pass filter, are used to process the data and various inferences are used to determine the activity like standing, sitting, lying, walking and falling accordingly. The system also provides a graphical user interface for data visualization, reporting and 3D real-time motion capturing. The architecture and implementation of the monitoring system are described. System testing results are also presented. |
