Falls detection and assessment

The paper has the objective to define a fall detection methodology using the sensors available on smartphone. Furthermore, the paper proposes a platform that support the remote fall detection methodology modeling the emergency management by using a collaborative approach. The advantage of the proposed platform is that the fall detection doesn't need of a new specific hardware as the proposed solution is based on a device such as a smartphone already widely used by senior citizens.In past studies and proposed systems the results of sensors applied on skin or on cloths showed better results than sensors equipping the smartphone. To overcome the low performances of former approaches using the accelerometer and, in some cases, the gyroscope equipping the smartphone the proposed methodology use five sensors equipping the smartphone: accelerometer, gyroscope, proximity, microphone and GPS. The results obtained using the five sensors are comparable to those obtained with skin and wearable sensors, but with the advantage to avoid the use of dedicated hardware.