A newly developed method to detect stealthy malware on IoT devices has been created by cybersecurity researchers, which uses electromagnetic field emanations to identify existing malware in every device. The method can also be applied to detecting obfuscated malware.
The recent discovery was introduced by researchers from IRISA (Research Institute of Computer Science and Random Systems) at a conference for computer security applications that happens once a year.
Threat actors utilise the side channel details to identify abnormalities in electromagnetic field emanations when they dissent from past patterns and suspicious behaviour in the system’s legal status.
With the help of the new method, cybersecurity tools will be able to detect ransomware, malware variants, or rootkits without any on-device alteration or customisation.
Furthermore, the electromagnetic field emanation weighed from the IoT devices is unobservable by malware. Therefore, unlike standard software monitoring and solutions, malware evasion techniques or obfuscations will be rendered useless.
Malware commonly does not have a feature to identify outside hardware solutions and a protection functionality against it even when it has full privileges from the device.
The new method operates in three phases: measuring electromagnetic field emanations when running 30 distinct malware variants and executing delicate activities to enhance a Convolutional Neural Network (CNN) model to classify hostile entities.
The framework also gets an executable as input and exports malware labels utilising the side-channel details. Utilising ordinary neural network models, researchers acquired better information about the state of a supervised IoT device.
The fast rate of growth and development in the emergence of IoT devices and appliances make them an attractive target for malicious threat actors. The cybercriminal environment is becoming more extensive, making monitoring, and identifying evasive malware more challenging. Researchers worldwide are expected to improve their malware analysis strategies to lessen security risks.
This newly developed method of using electromagnetic field emanation to identify malware will eventually catch the attention of numerous threat actors, resulting in the evolution of newer malware types.
