Abstract
Developing a passive monitoring prototype for elderly care is vital for ensuring their safety while respecting their autonomy. However, existing approaches often compromise confidentiality or necessitate the elderly to wear intrusive devices. To address this, we focus on recognizing behavioral patterns through Activities of Daily Living (ADLs). To detect and classify an ADL, data from various sensors is necessary, with motion sensors being the most prevalent. Consequently, our research delves deeper into understanding the functionality of these sensors and exploring the latest advancements in IoT communication protocols. Matter emerges as a standout choice, together with the OpenThread Border Router, which offer seamless interoperability, ensuring compatibility between devices and network infrastructure while prioritizing security and reliability. Leveraging the capabilities of these technologies, we developed a prototype for collecting data about an individual’s activity based on an occupancy sensor, found in a home setting. User-friendly management of the smart home is done through the Home Assistant platform, together with its HomeKit addon, enabling seamless interaction with commissioned devices directly from mobile phones. Our project advances through multiple stages, from laying the technical groundwork to defining data acquisition methods, successfully achieving these initial milestones. Our focus now shifts to data pre-processing, model development/selection for accurate sensor data classification within Activities of Daily Living (ADLs), and exploring system deployment options (cloud, edge, or on-device). Through our prototype, we showcase the feasibility of an elderly care monitoring system, emphasizing privacy and dignity with advanced IoT technologies.
Published Version
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