Pressure-Sensor System for Sleeping-Posture Classification

In this paper we present our research work on the classification of sleeping postures using pressure sensing. A densely pressure-sensor grid placed under the matrass is first used for the posture classification task that will serve next as a reference. Pressure data obtained from 21 subjects for 4 sleeping postures (supine, prone, left and right side) has been analysed to determine the optimal location and the number of pressure sensors for an efficient sleeping-posture classification system. Obtained settings are used to build the optimal system and a second experiment including 11 subjects has been carried out with the new system. The obtained results are very promising and show the potential and the interest of the proposed method compared to existing systems.