A new security system called LatentGesture, monitors users' touch patterns to prevent unauthorized access by locking the device when the touch signatures do not match.
Smartphone security is improving by the day with finger print scanners, password protection, or even pattern locks to protect the content stored on a smartphone or a tablet. Researchers from the Georgia Institute of Technology have found another method that outperforms the latest security systems in protecting mobile devices.
The new security system, LatentGesture, continuously monitors how a user taps and swipes to get to different apps and settings on a smartphone. The system captures the personal touch signature of the owner and uses that to validate third-party users. If the movements don't match with the owner's touch signature, the system accurately finds the mismatch and can be programmed to lock the device.
According to lead researcher Polo Chau, an assistant professor at the Georgia Tech College of Computing, people have a unique style of using a touchscreen. "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," Chau said, according to a report from Science Daily.
In order to test the system's accuracy, Chau and his team conducted a small experiment with 20 participants. Each participant was given a smartphone and tablet to fill out the given form by checking boxes and swiping slider bars. Since LatentGesture constantly runs in the background and monitors different touch patterns, the system was able to identify different participants and created a profile for each person.
Researchers assigned one participant as the owner of the device using the stored profiles and ran the tests again. LatentGesture identified owner's touch patterns and flagged everyone else as unauthorized users. The system was nearly 98 percent accurate on a smartphone and 97 percent accurate on tablets. Even by creating five profiles, one for the owner and other four for authorized users, the accuracy of the system was not compromised.
"It's pretty easy for someone to look over your shoulder while you're unlocking your phone and see your password," Samuel Clarke, another College of Computing student on the research team, said in a statement. "This system ensures security even if someone takes your phone or tablet and starts using it."
