BlueSeer: AI-Driven Environment Detection via BLE Scans*
TimeWednesday, July 13th3:30pm - 3:50pm PDT
Location3000, Level 3
Event Type
Research Manuscript
Design of Cyber-physical Systems, Cloud Computing and IoT
DescriptionIoT devices rely on environment detection to trigger environment-specific actions, e.g., for headphones to adapt noise cancellation. While phones feature many sensors, from GNSS to cameras, small wearables must rely on the few energy-efficient components they already incorporate. We demonstrate that a Bluetooth radio is the unique component required to accurately classify environments and present BlueSeer, an environment-detection system that solely relies on received BLE packets and an embedded neural network. BlueSeer achieves an accuracy of up to 82% differentiating between 7 environments on resource-constrained devices, and requires only 35 ms for inference on a 64 MHz CPU.