The world's largest companies are grappling with increasingly widespread and sophisticated malware attacks, but an interesting new detection technique could help them.
A team of researchers at France's Research Institute of Computer Science and Random Systems created an anti-malware system centered around aRaspberry Pi that scans devices for electromagnetic waves. According to Tom's Hardware, the security device uses an oscilloscope and H-Field probe to pick up anomalies in the waves that computers are under attack, a technique the researchers say is used to obtain precise knowledge.
CNN is used to determine if the data gathered indicates a threat. Researchers claim that they could record 100,000 measurement traces from internet of things devices that had been compromised by genuine samples and predicted three generic and one benign classes with an accuracy as high as 99.82%.
No software is required and the device you are scanning doesn't need to be manipulated. Bad actors will not be able to hide malicious code from detection software using obfuscation techniques.
Our method does not require any changes to the device. It can be deployed without any overhead. Researchers wrote in the paper that their approach can be easily detected and evaded by the authors.
The system was made for research purposes and not for release as a commercial product, so it may inspire security teams to look into novel ways of detecting malicious software. The neural network will need to be further trained before it could be used in practical ways.
The system is unique in that it makes it difficult for malicious people to hide their code, but it is not currently available to the public.
The other EM wave-scanning equipment costs thousands of dollars and is not a low-cost method for detecting malicious software. It is encouraging to see research approach such an important issue from a unique angle, who knows, this simple setup could one day help prevent the next major breach.
All Rights Reserved © 2024
