Engineers at Georgia Tech have created tiny, battery-free metal tags that produce a short ultrasonic pulse when struck. The tags are metal disks with a central hole and cutouts around the edge; mounted on a small 3D-printed base, they attach to cabinet doors, doorframes, faucets or other surfaces. When a tab on a moving part hits a disk, the impact generates an ultrasonic pulse that humans cannot hear but that a nearby wearable microphone can detect and log.
The exact shape of each disk determines its resonant frequency, so each tag produces a distinct acoustic "fingerprint" that can be used for activity recognition, as Yibo Fu explains. Possible applications include counting repetitions on gym equipment, monitoring faucets or toilet lids to track water use or signal that an elderly person may need help, and button-style tags to start timers or log actions. A video Fu posted on Instagram went viral, drawing about 1.6 million views, 150,000 likes and hundreds of comments, with viewers suggesting other uses.
Bolei Deng led the vibration modeling and simulation work. The team’s simulations produced nearly 1,300 initial designs that each generate a unique ultrasonic frequency above 20 kilohertz, and the researchers tested 15 designs. With careful design the set could expand to dozens, hundreds or even thousands of distinct tags. Because ultrasound is easy to detect in noisy environments and does not travel far, only nearby microphones hear a tag. The team avoided complex machine learning and instead used a simple hardcoded algorithm, which reduces computational and electrical demands. The project was a collaboration among engineering and computing researchers and appeared in the Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies.
Difficult words
- ultrasonic — sound at frequencies above human hearing range
- resonant frequency — natural frequency at which an object vibrates
- wearable — electronic device designed to be worn
- simulation — computer model that imitates real behaviorsimulations
- hardcoded algorithm — fixed set of program rules, not learned
- battery-free — operates without needing an internal battery
Tip: hover, focus or tap highlighted words in the article to see quick definitions while you read or listen.
Discussion questions
- Which everyday uses of these battery-free ultrasonic tags seem most useful to you, and why?
- What privacy or safety issues could arise if many people wore microphones that detect these tags?
- How might using a hardcoded algorithm instead of machine learning affect the devices' long-term flexibility and cost?
Related articles
AI and Wearable Devices for Type 2 Diabetes
A meta-review from the University at Buffalo examines AI-enhanced wearable devices for Type 2 diabetes and prediabetes. The study finds predictive benefits and important limits, and calls for larger, more transparent studies before routine clinical use.
AI tool helps people understand autistic communication
Researchers created NeuroBridge, an AI tool that shows how autistic and non-autistic people can interpret speech differently. The tool gives example replies and trains clearer, more direct communication based on autistic preferences.
Reducing unsafe responses in large language models
Researchers studied how large language models (LLMs) handle safety and tested training methods to reduce unsafe outputs while keeping performance. They identified key challenges and a technique that preserves safety during fine-tuning.
