Abstract

Many residential environments have been equipped with sensing technologies both to provide assistance to older people who have opted for aging-in-place and to provide information to caregivers and family. However, such technologies are often accompanied by physical discomfort, privacy concerns, and complexity of use. We explored the feasibility of monitoring home activity using chemical sensors that pose fewer privacy concerns than, for example, video-cameras and which do not suffer from blind spots. We built a monitoring device that integrates a sensor array and IoT capabilities to gather the necessary data about a resident in his/her living space. Over a period of 3 months, we uninterruptedly measured the living space of a typical elder person living on his/her own. To record the level of activity during the same period and obtain a ground truth for the activity, a set of motion sensors were also deployed in the house. Home activity was extracted from a PCA space moving-window which translated sensor data into the event space; this also compensated for environmental and sensor drift. Our results show that it is possible to monitor the person’s home activity and detect sudden deviations from it using a low-cost, non-invasive, system based on gas sensors that gather data on the air composition in the living space. We made the dataset publicly available at a data repository https://doi.org/10.24432/C5762W.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call