Floor Vibration-based fall detectors

From Capsil Wiki

Floor vibration solutions falls into two categories. One type is based on the use of acoustic analysis of the sounds made when a person falls a contacts a solid object or surface such a floor. However there are limitations to the technique especially with composite surfaces such a concrete with carpet overlay for example.

The second type uses conversion of the mechanical energy that is generated when a person impacts an object/surface such as a floor it is converted into electrical energy using piezoelectric sensors. The selectivity of the sensors is based on the hypothesis that the vibration signature of the floor generated by a fall is significantly different from those generated by normal daily activities such as walking etc. and secondly the vibration signature of a human fall is significantly different from those of falling objects. The Medical Automation Research Centre (MARC) at the University of Virginia have demonstrated a sensor based on this approach. In controlled experiments conducted to test this fall detector they attained 100%true positives and 0% false alarms with a detection range of 20 feet.

References

[1]