VibraPhone attack turns smartphones into listening devices. Two researchers from the University of Illinois at Urbana-Champaign have devised a method for turning vibration motors, like the ones found in smartphones, into makeshift microphones, capable of recording the sound around them.
Their method doesn’t yield perfect results and also needs physical access to the device, but it puts in place the theoretical details needed to carry out and refine such attacks in the future.
The attack, named VibraPhone, is based on the idea that any vibration motor is technically a speaker. Vibration motors translate electrical current sent into sound waves by moving a coil. In this case, the coil generates vibrations and low humming sounds in the phone.
Since a microphone is basically a reversed speaker, taking incoming sounds and converting them back into electrical waves, the researchers decided to attempt an experiment during which they turned a phone’s vibration motor into a microphone.
VibraPhone requires attackers to rewire the phone’s vibration motor
The researchers say they had to rewire the phone’s vibration motor, which makes this highly unlikely to be an attack usable in the real world unless the attacker has enough time to break the phone apart and rewire its vibration motor.
In the experiment attempted by the researchers, they said they were able to record sounds, but that the sound’s quality was very poor.
This was because the vibration motor was not optimized for recording audio and was able to record sound waves of maximum frequencies of up to 2 kHz, the lower end of the spectrum, leaving out high-pitch noises.
Rewiring a smartphone for a VibraPhone attack
Some audio signal rebuilding was needed
The researchers applied various algorithms to the recorded sound wave in order to optimize its output and attempt to fill in the missing audio waves corresponding to the high-frequency sound.
The results weren’t optimal, but four in five persons were able to understand spoken words. The researchers did not attempt to refine this algorithm and said that someone that has an interest in optimizing sound pick-up via vibration motors would be able to get better results with more effort.
“Even automatic speech recognizers were able to decode the majority of the detected words and phrases, especially at higher loudness,” the researchers said. “The application space of such systems remains open, and could range from malware eavesdropping into human phone conversation, to voice controlled wearables, to better microphones that use the vibra-motor as a second MIMO-antenna.”
Below is a video explaining the attack. More details and examples are available via the VibraPhone website, and in the Listening through a Vibration Motor research paper by Nirupam Roy and Romit Roy Choudhury.