Researchers at the Massachusetts Institute of Technology’s Computer Science and Artificial Intelligence Lab (CSAIL) have created a way of detecting emotions by using wireless signals.
“All the time, you’re surrounded by wireless signals from WiFi, TV, and other such things,” Ph.D. researcher Fadel Adib told Digital Trends. “These signals interact with the human body, so as you move, the waves that are in the environment change with you — a bit like water swirling around you when you’re in a swimming pool. By extracting and analyzing these signals and their interactions with individuals, we can get very precise details about people’s breathing and heartbeat.”
The idea behind the work is that current technologies for emotion tracking, like facial-recognition cameras that look for certain expressions, are not always reliable — as anyone who has ever employed a poker face during a game of cards will know. By using the so-called EQ-Radio system, however, it is possible to establish whether a person is excited, happy, angry, or sad with 87 percent accuracy.
To place it into further context, by analyzing wireless signals reflected off people’s bodies, EQ-Radio is able to measure heartbeats with the same accuracy as an ECG monitor used in cardiology tests.
The work is due to be presented by Adib, MIT professor Dina Katabi, and fellow Ph.D. student Mingmin Zhao at the Association of Computing Machinery’s International Conference on Mobile Computing and Networking (MobiCom) in October. There are also plans to spin some aspects of the technology off into a commercial company.
“Some of the applications this could be applied to include smart environments,” Adib said. “For example, if you come back home at the end of a long day, your home could use the wireless signals to recognize your emotions. If it knows that you’re tired, it would be possible to choose some relaxing music or dim the lights.”
Other potential applications may be as diagnostic technology to tell if people are depressed, or possibly even as an artistic tool to help movie makers or advertisers gauge the emotional reactions of viewers.
“I’m really excited about this work,” Adib said. EQ-Radio can probably tell.
- Smart toys accurately predict whether a guide dog will make it through training
- Sonos vs. Bluesound: A Hi-fi, Wi-Fi Speaker System Shootout
- The weirdest, wildest, and most wonderful wearables in the world
- The mind-reading AlterEgo headset (almost) promises telepathy with Alexa
- Can this app make EarPods sound like studio-grade cans? We tested to find out