Integration of Multiple Sensor-Based Wellness Systems for Supportive Housing Using Bluetooth Low Energy

Residents in permanent supportive housing, defined as long-term, community-based housing combined with supportive services, often have health conditions requiring daily medication and monitoring for disabling incidents inside their home. Sensor-based solutions for checking for immobile residents, and automatic pill dispensers to ensure the timely delivery of medications are being developed to support living with health conditions. This work describes how multiple such health monitoring devices can be coordinated from a single user interface with communication over Bluetooth Low Energy (BLE). Individual subsystems, including an automatic pill dispenser and a smart wellness check, are designed for use within a supportive housing apartment unit. An Arduino RP2040 micro controller is used as each subsystem's central processing unit. The foundation of the communication between the different subsystems is the BLE protocol. BLE was chosen for its adaptability to heterogeneous devices and low power consumption. The BLE communication architecture links these devices, facilitating reliable data transfer and device-to-device coordination. An Android application running on a tablet PC acts as the central coordinator of all subsystems and provides a unified user interface. The application scans for BLE devices, connects to discovered subsystems, and displays real-time sensor information in a user-friendly format. Nearby peripheral devices are identified using BLE advertising and scanning. Bidirectional communication channels are created once linked, enabling the central device to communicate with the peripherals and receive data from them. This BLE architecture supports simultaneous communication with multiple peripherals. Different parameters of the BLE protocol are systematically varied; the parameters are the MTU size, time interval for BLE device scans, and the resulting number of GATT connections and disconnections under various time intervals. Fine tuning these parameters enables the identification of the optimal set of parameter values for reliable data transfer between the central user interface and individual subsystems with minimum latency and maximum energy conservation.