With the next wave in consumer electronics expected to be smart bracelets, watches, and eyewear, experts are hopeful that thought-based authentication—which recent research shows is possible—could have a significant impact on the technology.
Wearable devices will monitor everything from exercise routines to sleep quality and vital signs. (See also "Interactive Apparel: Are Those Pants, or Is That a Keyboard You're Wearing?")Because they are so close to the body, embedding sensors that read brainwaves to identify the wearer would be an easy and unobtrusive means of authentication, said John Chuang, lead researcher in the University of California, Berkeley, study on thought-based authentication.
"Wearable computing devices is a natural candidate for integrating with this type of technology," Chuang said.
First step: Identifying brainwave action
While that may be true, thought-based authentication is years away from becoming mainstream. What Chuang and his colleagues did was to achieve 99 percent accuracy in identifying the unique brainwaves of 15 undergraduate and graduate students who participated in the study.
Just as important, the researchers were able to do it using an off-the-shelf gaming headset that captured electro-encephalography (EEG) signals.
Trent Henry, an analyst with Gartner, said the results were "interesting" and justified further study. "It's certainly exciting to warrant some additional research, but we're a long, long way from market viability for this kind of technology," Henry said.
Biometrics has been studied for sometime as a way to replace or complement user names and passwords. Technology exists today for reading fingerprints, scanning retinas, and recognizing a person's voice.
In the U.C. study, each of the subjects used a Neurosky Mindset that places a single dry-contact sensor on the forehead. Other than the sensor, the device looks the same as any Bluetooth headset used with mobile phones, music players and other computing devices. The Mindset costs about $100.
The headset is much simpler than EEG systems used in clinical environments. Those systems employ a dense array of electrodes placed on a person's head to provide as many as 256 channels of data.
Testing repetitive tasks
In the U.C. experiment, the subjects repeated each of seven tasks five times during two sessions, providing researchers with a total of 1050 brainwave data samples.
The tasks included:
- Closing the eyes and focusing on breathing for ten seconds.
- Selecting a repetitive motion from a sport and imagine moving the body to perform the motion.
- Silently singing a song or reciting a passage.
- Opening their eyes after hearing an audio tone and staring at a dot on a piece of paper for five seconds.
- Silently counting the number of boxes in a color of their choosing among a grid of colored boxes.
- Focusing on a mental thought of their choosing for ten seconds.
Researchers found that the EEG signals exhibited patterns that were unique to the individuals. They also found that the mental task performed did not matter in terms of producing identifying patterns. However, people preferred tasks in which they chose the secret color, thought, song, or passage.
On a practical level, the study has its weaknesses. Having to focus on a thought for five to ten seconds would seem like an eternity in computing, which people expect to deliver instantaneous responses. In addition, the study did not investigate whether an impersonator could successfully use someone's "pass-thought."
Nevertheless, the experiment moves brainwave authentication forward, and shows that the technology is deserving of further study. (See also Wearable Technology: 10 Gadgets Available Now.")
This story, "Study shows brainwaves could be security for wearable tech" was originally published by CSO.