Rutgers develops tech that can turn any surface into a touch screen
The future is now
Rutgers scientists have developed new technology that effectively turns any solid surface into a touch-sensitive panel. It will allow the user to unlock their front door, car or any other lock by drawing a pattern or gesture on the surface.
The device is called VibWrite because it works by transmitting high-frequency vibrations through the surface it is mounted on, said Jian Liu, a Ph.D. student who is part of the team that has been working on the project.
Yingying Chen, a professor in the Department of Electrical and Computer Engineering, is responsible for overseeing the project and said in an email that VibWrite is based on entirely new research. It is the first device to use physical vibration as a means for user authentication.
VibWrite is comprised of a motor, which generates physical vibrations, and a receiver, which detects any changes in the vibrations, Liu said. If some object, such as a finger, touches the surface between the motor and the receiver, the device recognizes the change in vibrations and so the surface functions like a touchscreen.
Because VibWrite is based on such simple machinery, it will be less expensive than similar high-tech authentication systems, he said. Parts used in current prototype-builds of the system costs less than $50 in total.
“We can enable any solid surface, like a door panel, a table, a blackboard, any solid surface to have the ability to sense,” Liu said. “We just attach a motor and a sensor onto the surface so there is a sensing area, and that surface has the ability to sense. The sensor is tiny and easy to deploy.”
He said the vibrations created by the motor are of such high frequency and low amplitude that a person touching the surface will not feel any difference.
Because every person's bone structure is unique, no two people will change the vibrations in exactly the same way. So in theory, VibWrite cannot be hacked, Liu said.
“For a smartphone, a touchscreen, if you know the user's password or lock pattern, then you can directly hack the system,” said Chen Wang, another Ph.D. student working on the project. “But for our system, even if you know the credentials you still cannot pass it because we also consider the user's physiological characteristics. Different people interfere with the vibrations differently, and they cause the receiver to receive different signals.”
Liu said the current build of VibWrite can be programmed for three kinds of securities: a PIN number, which is input by touch on a three-by-three gird where each space is assigned a digit from one to nine, a lock pattern, which the user draws with his finger on a three-by-three gird or a “flexible gesture,” which is any shape or gesture that the user may draw on the surface. The team's goal is for a user to unlock the system by drawing his signature.
VibWrite is still in the prototype stage. In experiments, the system recognizes the legitimate user's first attempt over 90 percent of the time, Liu said. But it is not perfect yet. If an attacker's finger is similar enough to the legitimate user's, the attacker can occasionally pass the system.
Wang described the most challenging attack model, in which the attacker knows the password and the exact way that the legitimate user performs it — the speed, the location on the sensitive zone — and mimics it perfectly. Before VibWrite is made available for consumers, it must be able to reject such attackers 100 percent of the time, he said.
Chen said VibWrite will be commercialized after its accuracy is improved. She expects it to be available to the public in a few years.
“The current system still needs further improvement to cope with many practical environmental factors related to the solid surface that may affect the system performance ... temperature, humidity, wind, wetness, dirt,” Chen said. “We also need additional efforts to further improve the accuracy of the current VibWrite system. To improve performance, we will deploy multiple sensor pairs, refine the hardware and upgrade authentication algorithms.”