Cybersecurity researchers from the Georgia Institute of Technology have developed a new security system that can identify individual smartphone or tablet owners simply by how they touch and swipe the screen.
Called LatentGesture, it builds up a touch profile or signature of the user and then uses that information to constantly check if the person using a device is tapping and swiping it in the same way. If it detects differences, it can lock the user out of the device.
"Just like your fingerprint, everyone is unique when they use a touchscreen," said Polo Chau, a Georgia Tech College of Computing assistant professor who led the study.
"Some people slide the bar with one quick swipe. Others gradually move it across the screen. Everyone taps the screen with different pressures while checking boxes."
In tests LatentGesture was nearly 98 % accurate on an Android smartphone and scored 97 % on a tablet. What's more, because the system can recognize and store several signatures, LatentGesture could be a great way of setting up a device for family use.
"This feature could be used when a child uses her dad's tablet," said College of Computing sophomore Premkumar Saravanan. "The system would recognize her touch signature and allow her to use the device. But if she tried to buy an app, the system could prevent it."
The study will be formally presented in Toronto at ACM Chinese CHI 2014 from April 26 to 27, and while LatentGesture is potentially a brilliantly simple solution to the growing problem of securing digital devices and digital content, it is by no means the only such system currently in development.
In September, a group of scientists at the Illinois Institute of Technology, led by Cheng Bo, showcased a system called SilentSense.
It uses a smartphone's existing sensors, such as its gyroscope and accelerometer, to record how the device's owner uses his or her phone. The amount of pressure they exert, where, and for how long, when touching or swiping the screen, the size of the owner's finger, and favored holding angles and use positions are used to build up a biometric image of the owner. It can even recognize and identify a user's gait from movement and direction detected by the sensors when the owner is walking.
"Different users, dependent on sex and age among other things, will have different habits in interacting," says Bo of the system. Indeed, it is so sensitive that it is able to identify a new user with 98 % accuracy after a mere 2.3 touches. That figure moves to 99 % after 10 taps.
But like LatentGesture, SilentSense is still in the development and testing stage, meaning that for the time being at least, consumers who want the extra reassurance of a biometric security system for protecting their smartphone will have to make do with the fingerprint scanners on the Apple iPhone 5S or the Samsung Galaxy S5.