In latest research, mobile phone calls are remotely intercepted via sensors

In latest research, mobile phone calls are remotely intercepted via sensors ...

Researchers have demonstrated a technique to detect the vibrations of a mobile phone''s earpiece and decipher what the person on the other side of the call was saying with up to 83% accuracy. The researchers at Pennsylvania State University used an off-the-shelf automotive radar sensor and a novel processing technique to identify this major security danger.

"The misuse of such sensing techniques by adversaries becomes probable," said Penn State''s Suryoday Basak.

"Our demonstration of this type of exploitation contributes to the slew of scientific literature that says, "Hey! Automotive radars can be used to intercept audio. We need to do something," Basak said.

The radar is used in the millimetre-wave (mmWave) spectrum, particularly in the bands of 60 to 64GHz and 77 to 81GHz, which led the researchers to name their approach "mmSpy." This is a subset of the radio spectrum used for 5G, the fifth-generation standard for communication systems across the globe.

Researchers simulated people speaking through the earpiece of a smartphone in the mmSpy speech described in the IEEE Symposium on Security and Privacy (SP) in 2022.

The earpiece of the phone vibrates from the speech, and the vibration spreads across the phone''s body.

  • Meta Unveils ''''Make-A-Video'''' AI Text-To-Video Generator: Details

"We use the radar to sense this vibration and reconstruct what was said by the person on the other side of the line," said Basak.

Mahanth Gowda, an assistant professor at Penn State, stated that their approach works even if the audio is completely unaudible to both humans and microphones nearby.

"This isn''t the first time similar vulnerabilities or attack methods have been discovered," Basak said. This particular aspect of understanding and reconstructing speech from the other side of a smartphone line has not been explored.

The researchers then send them to machine learning modules designed to classify speech and reconstruct audio.

According to reports, when the radar detects vibrations from a foot away, the processed speech is 83 percent accuracy. That drops as the radar moves from the phone, down to 43 percent accuracy at six feet.

As the speech is restored, researchers may then filter, enhance, or classify keywords as needed, according to Basak.

The team is continuing to develop their approach in order to better understand non only how to protect against this security danger, but also how to exploit it for the future.

"The methodology that we developed can also be used to detect vibrations in industrial equipment, smart home systems, and building-monitoring systems," Basak said.

According to the researchers, similar home maintenance or even health monitoring systems may benefit from such sensitive monitoring.

"Imagine a radar that might track a user and solicit assistance if some health parameter changes in a dangerous way," Basak said.

"With the right set of target actions, radars in smart homes and industries can enable a swift turnaround when problems and issues are identified, according to Andrew Goldberg.