For those not in the know, a Raspberry Pi is a nifty little single-board computer used for a staggering number of projects. If your project needs a computer or controller of some kind, chances are a Pi is the answer.
That must’ve been the mentality the researchers at the Research Institute of Computer Science and Random Systems (IRISA) in France possessed. The team of Annelie Heuser, Matthieu Mastio, Duy-Phuc Pham, and Damien Marion turned a Pi into a new malware detection system. For the technically minded, you can read their paper here. For the rest, we’ll do our best to give you the gist of it.
This malware detection system doesn’t use software like other systems. It uses an external oscilloscope and an H-field probe (you don’t have these at home). These scan and detect specific electromagnetic (ME) waves that indicate malware on an affected device.
This Pi eats malware
Doing it this way, the research team was able to “obtain precise knowledge about malware type and identity” from the scans.
Current malware detection requires additional software installed on the compromised system. Think antivirus software like ESET or Malwarebytes. Often, the malware is advanced enough to exploit or bypass the very software you’re using to remove it.
By avoiding the need for software, the standard methods used by malware writers to hide their code just don’t work.
The Pi used in this project used Convolution Neural Networks (CNN) to scan their data sets for malware. This led to an effective detection rate of up to 99.82%.
As impressive as that is, you won’t find one of these tests on any nearby shelf. That’s a pity. If this sort of testing becomes widely available, malware coders will need to seriously up their game. It can’t be that hard to think of ways to hide their code’s EM waves. Can it?