A noncontact-sensor surveillance system towards assisting independent living for older people

In this paper, we propose a framework which provides the temporal and spatial context for daily routine inferring in an ambient-assisted-living environment equipped with several sets of passive infrared (PIR) sensors. The inferred information could help identify the user's daily pattern and abnormalities from long-term monitoring. We implemented the prototype of the system from the electric parts and software parts. The effectiveness of the proposed framework is reflected by the initial results over one day's continuous monitoring of two individual older participants. The framework could satisfy the basic needs of older people who seek an unobtrusive and practical way to assist their daily life. The extended system can be expected to combine with wearable sensors to meet different applications.