The Internet of Things (IoT) is a network of interconnected devices, becoming increasingly complicated and suffering from inadequate security measures. Cybercriminals, especially those who specialise in malware and rootkits, recently target them because they often use outdated technology without taking security risks into account.

In this session, we will discuss two challenges: rootkit detection and malware classification in the help of leveraging electromagnetic (EM) side channels. EM allows us to operate outside of the “box” (literary device), with no resource requirement on the target device. Our approach focuses on the ARM and MIPS architectures of Raspberry Pi and Creator CI20 real-world devices. The solution employs multiple data preprocessing methods, allowing analysts to select a variety of machine learning and deep learning models based on their specific requirements. Both approaches resulted in high accuracy (upto 100%) for multiple malware classification and real-time detection scenarios.