Context-aware smart home monitoring through pressure measurement sequences

Abstract

This paper presents the architecture, sensing and results of a context-aware smart home monitoring system based on pressure measurement sequences. It focuses on the analysis of transfers performed by the occupant in the bedroom and bathroom to assess if their behavior is within a normal range of motion. Pressure sensors are placed under the bed mattress and embedded in the grab bars of a toilet commode to collect data sequences in the form of different modalities. From these sequences, relevant clinical features are extracted and fused with both past and expected results to output a warning level. The data fusion outputs are then fed into a simulated context-aware classifier before a final decision regarding the occupant is made. The pressure sensing system can be modified depending on a particular occupant's needs, and can also fit well within a larger and more complex smart monitoring framework.