Watch out: If you've got a smartwatch, hackers could get your data

ECE News

David Robertson, CSL
9/10/2015

Story Highlights

  • Associate Professor Romit Roy Choudhury's group has found smartwatch sensors could cause privacy concerns for their users.
  • Using a homegrown app on a Samsung Gear Live smartwatch, the researchers were able to guess what a user was typing through data "leaks" produced by the watches' motion sensors.
  • The work, funded by the National Science Foundation, is being presented this week at the MobiCom 2015 conference in Paris.

They’re the latest rage in jewelry and gadgetry, but like all computer devices, smartwatches are vulnerable to hackers, say ECE ILLINOIS researchers.

Romit  Roy Choudhury
Romit Roy Choudhury
Using a homegrown app on a Samsung Gear Live smartwatch, the researchers were able to guess what a user was typing through data “leaks” produced by the watches' motion sensors.

The project, called Motion Leaks through Smartwatch Sensors, or MoLe, has privacy implications, as an app that is camouflaged as a pedometer, for example, could gather data from emails, search queries, and other confidential documents.

The work, funded by the National Science Foundation, is being presented this week at the MobiCom 2015 conference in Paris.

"Sensor data from wearable devices will clearly be a double-edged sword,” said Associate Professor Romit Roy Choudhury, who is affiliated with the Coordinated Science Laboratory. “While the device's contact to the human body will offer invaluable insights into human health and context, it will also make way for deeper violation into human privacy. The core challenge is in characterizing what can or cannot be inferred from sensor data and the MoLe project is one example along this direction.”

Hackers could access data produced by motion sensors on a smartwatch and use it to guess what a user is typing.
Hackers could access data produced by motion sensors on a smartwatch and use it to guess what a user is typing.
The app uses an accelerometer and gyroscope to track the micro-motion of keystrokes as a wearer types on a keyboard. After collecting the sensor data, researchers ran it through a “Keystroke Detection” module, which analyzed the timing of each keystroke and the net 2-D displacement of the watch. For example, the left wrist moves farther to type a “T” than an “F.” You can see how it works in the researchers' video.

While Illinois researchers developed MoLe, it is conceivable that hackers could build a similar app and deploy it to iTunes and other libraries.

Roy Choudhury’s team said the rapid proliferation of wearable devices made them ask the question: Just how secure is the data? They approached this topic from the perspective of an attacker. Rather than directly developing security measures for smartwatches, they aimed to discern ways that attackers can decipher users’ information.

“There are a lot of good things that smartwatches can bring to our lives, but there could be bad things,” said graduate student He Wang. “So if you think from that perspective  — if there are any ‘bad’ things we could do — we can help other people protect their privacy, or at least make them realize there’s a potential problem.”

A possible solution to these motion leaks would be to lower the sample rate of the sensors in the watch, Wang said. For instance, the sample rate is normally around 200 Hertz, meaning the system logs 200 accelerometer and gyroscope readings per second. However, if that number is lowered to below 15, the users’ wrist movements becomes extremely difficult to track.

While their work has yielded revolutionary results so far, there is still a long way to go in polishing the data-collection process. The team’s current system can’t detect special characters such as numbers, punctuation, and symbols that might appear in passwords. The space bar or key also poses an obstacle. In addition, researchers can only collect data from the hand wearing the watch and from people who have standard typing patterns.

“There’s a subset of people who don’t type like that,” said Ted Tsung-Te Lai, a postdoctoratal researcher at Illinois, who noted the team will develop more models to account for typing differences in the future.

While a Samsung watch was used in this project, the researchers believe that any wearable device that uses motion sensors – from the Apple Watch to Fitbit – could be vulnerable as well.

"We would just like to advise people who use the watch to enjoy it," Lai said, “but know, ‘Hey, there’s a threat’.”

Media Contact

Julia Sullivan

Assistant Director of Communications
1064 ECE Building
(217) 300-3731
juliams@illinois.edu

Todd Sweet

Director of Communications
1066 ECE Building
(217) 333-5943
tmsweet@illinois.edu