Bluetooth device for proximity sensing
Most proximity datasets come from smartphones, but phone-based sensing is inconsistent (different hardware, placements, background behaviors) and hard to standardize for research. This project explores a dedicated BLE wearable that can reliably detect close-contact events and generate clean logs for analyzing contact networks and disease transmission dynamics.
The system uses BLE advertising + scanning on nRF52840 devices to detect nearby wearables and record RSSI over time. Those raw signals are then processed into stable “meeting” events (who met whom, for how long, how often), creating a dataset that’s useful for outbreak-style modeling and controlled campus experiments.
Rather than treating RSSI as perfect distance, the focus is robust contact detection: filtering noise, reducing flicker near thresholds, and producing event logs that are consistent enough to support downstream epidemiological analysis.