Key Takeaways
- University of Osaka researchers built a wireless EEG transmitter that operates entirely on body heat using a thermoelectric generator.
- The system uses compressed sensing to undersample EEG signals, then reconstructs them on the receiver side with high accuracy.
- It achieved a compression ratio of six while maintaining an average normalised mean square error of just 0.047 on the CHB-MIT EEG dataset.
- During an outdoor demo at Expo 2025, the device ran continuously at temperatures above 32 degrees Celsius with no airflow or backup power.
- Lead author Daisuke Kanemoto said the long term goal is “sensing systems that can operate indefinitely without maintenance.”
Wearable health devices have long been hamstrung by their need for power. Batteries add weight, require recharging, and limit how long sensors can operate, especially for continuous monitoring outside clinical settings. A team at the University of Osaka has just shown that for at least one important use case, monitoring brainwaves, batteries may not be necessary at all.
Demonstration experiment at Expo 2025.
(Although a fan is visible in the filming environment, it was not used in the experiment, which was conducted at ambient temperatures in the 32°C range.)
Original content, No modifications of the work are permitted., Daisuke Kanemoto
The system, presented at the IEEE International Conference on Consumer Electronics and demonstrated live at Expo 2025 in Osaka, captures small amounts of energy from the temperature difference between the wearer’s skin and ambient air using a thermoelectric generator. To stretch that tiny power budget, the engineers turned to compressed sensing, randomly sampling only a fraction of the EEG signal and using a reconstruction algorithm to rebuild the full waveform on the receiving end. The approach hit a compression ratio of six while keeping the reconstruction error remarkably low.
What makes the demo especially notable is that it worked outdoors in warm conditions. Even at temperatures above 32 degrees Celsius, with only a few degrees of difference between body and air, the device sustained continuous wireless EEG transmission. Lead author Daisuke Kanemoto said the work is an important step toward “practical, maintenance free sensing technologies.” Beyond brain monitoring, the same architecture could power infrastructure sensors, environmental monitoring, and smart city deployments where battery replacement is impractical.
Your First 10 AI Skills
10 practical AI skills, copy-paste prompts and a 7-day plan to start using AI with confidence.
Download the guide →