Deus EM Machina: On-Touch Contextual Functionality for Smart IoT Appliances
summary
A smartphone augmented with a copper foil antenna and a $4 microcontroller samples electromagnetic noise at up to 2 MHz, extracts FFT features, and runs an on‑device SVM classifier to identify touched appliances. The system trains on 20,000 signatures from 17 appliances and achieves 98.8% recognition accuracy in a 10‑user study. The team ships a USB sensor prototype, implements libcharm for contextual verbs, and demonstrates 12 example applications, six fully functional end‑to‑end.
AI-written summary · about
chapters (5)
related talks
29mSmartAuth: User-Centered Authorization for the Internet of Things
23mA Manufacturer's Post-Shipment Approach to Fend-Off IoT Malware in Home Appliances
40mA Co-Design approach for automated cooking appliances
58mInternet of things Barcodes, QRs, NFC, EPC & Arduino
51mThe Connected Smart Home, from IoT to Cloud
07mInterview
49m40875 The dark Age of IoT
51mThings That Go Bump in the Night: Detecting Problems in the Internet of Things