For malware detection, the Raspberry Pi has been used in conjunction with other gear.

The system was employed by the researchers without the need of any special software.

Malware attacks have increased dramatically in recent months.

A group of academics has discovered that the Raspberry Pi, a single-board computer popular for powering DIY projects, can now identify malware using electromagnetic waves. The newly designed technology is said to be capable of detecting malware without the use of any additional software.

In early tests, the Raspberry Pi-equipped hardware was able to detect malware with nearly 100% accuracy, according to the researchers. Small and medium-sized businesses might use the development to help secure their systems against cyberattacks at a reasonable cost if it becomes commercially available. 

The malware detection system was developed using Raspberry Pi by a team of researchers from the Research Institute of Computer Science and Random Systems (IRISA) in France, which included Annelie Heuser, Matthieu Mastio, Duy-Phuc Pham, and Damien Marion.

According to Tom's Hardware, the researchers used an oscilloscope (Picoscope 6407) and an H-Field probe, as well as a Raspberry Pi 2B, to scan devices for certain electromagnetic waves that can indicate whether malware is present. 

The researchers employed Convolution Neural Networks (CNN) to evaluate the data for malware dangers, according to a study article published last month. 'Our technique does not necessitate any changes to the target device.' As a result, it may be deployed without any overhead, regardless of the resources available. Furthermore, our technique has the advantage of being difficult for malware authors to identify and avoid,' the researchers noted in the article.

The scientists stated that they were capable of recording 100,000 measurement traces from an Internet of Things (IoT) device afflicted with various in-the-wild malware strains as well as realistic benign behaviour using their reference design. The team also claimed to have predicted three generic malware classes and one benign class with a 99.82 percent accuracy.

To get around software-level malware detection, hackers frequently use obfuscation techniques. However, because the new model does not rely on software to detect threats and instead relies solely on hardware and electromagnetic waves, it may be able to analyse and detect malware that was previously undetected by specialised softwares. It's vital to emphasise that the researchers' system was created just for research reasons and is not intended for commercial use. Manufacturers may be prompted to develop a standalone solution that uses electromagnetic waves to identify malware and other risks in the future.

Due to the COVID-19 restrictions, people began utilising the Internet more in the last year, which increased the number of cyberattacks. According to a recent analysis by Check Point, weekly cyberattacks have climbed by 50% since 2020. According to the cybersecurity firm, cyberattacks in India climbed by 24% year over year to 1,830 weekly attacks per organisation in 2021.