A wearable inertial sensing device for fall detection and motion tracking

Fall is a major problem for the elderly persons and patients suffering from neuromuscular disorders. A body-worn fall detector can greatly help in providing timely assistance by alerting the emergency services and relatives in case of such an occurrence. A wearable inertial sensing device is developed for fall detection and motion tracking. It consists of an integrated sensor with tri-axial accelerometer and tri-axial gyroscope for sensing linear acceleration and angular velocity, a microcontroller for data acquisition and processing, nonvolatile memory for recording the sensed data, and Bluetooth for wireless connectivity. A fall detection algorithm applying multiple decompositions and thresholding on the acceleration data is implemented for real-time processing using sampling frequency of 100 Hz. The sensed data can be continuously recorded for approximately 10 hours at a sampling frequency of 20 Hz and therefore the device can also be used for actigraphy for diagnosis of sleep disorders. A number of such devices worn on the different parts of the body and wirelessly connected to a central device can be used for continuous tracking of the posture and movement and can serve as an aid for assisted living.