Doctor of Philosophy, Case Western Reserve University, 2010, EECS - Electrical Engineering
A real-time method using only accelerometers is developed for classification of basic static/dynamic human postures, namely sitting, standing, bending, walking, lying, and running, as well as the dynamic states between them. Discrete wavelet transform (DWT) in combination with a fuzzy logic inference system (FIS) are the algorithmic basis underlying this method.
A generic platform for continuously and unobtrusively monitoring human motion activity and safety is developed that is low power, inexpensive, and wearable. The platform is built around the following key components: a commercial low-power 10-million-instruction-per-second (MIPS) microcontroller; an IEEE 802.15.4 compliant 2.4 GHz wireless transceiver; sensors, including accelerometers, microphone, and humidity/temperature sensors. The sampling frequency is in the range of 20-100 Hz. The hardware architecture is a distributed modular implementation, occupying an area of less than one square inch. The hardware is integrated in a conventional wearable headband.
Wirelessly transmitted data from a single three-axis accelerometer integrated into the headband is collected in real time on a laptop, and then analyzed to extract two sets of features necessary for posture/movement classification. The received acceleration signals is decomposed with DWT to extract the first set of features; any change of the smoothness of the signal that reflects a transition between postures is detected at the finer DWT resolution levels. FIS then uses the previous posture transition and the second set of features to choose one of eight different posture categories, namely sit, stand, lie on back, lie on left, lie on right, bend, walk, and run. Using the classifier in typical everyday activity among multiple users indicated more than 96.9%, 94.2%, 97.5% accuracy in detecting the static postures, walking, and running, respectively. Identifying the dynamic transitions among these steady postures achieved 92.6% accuracy.
Furthermore, a (open full item for complete abstract)
Committee: Mehran Mehregany PhD (Committee Chair); Kenneth Loparo PhD (Committee Member); Frank Merat PhD (Committee Member); Joseph Mansour PhD (Committee Member)
Subjects: Electrical Engineering