ROBUST MALWARE DETECTION FOR INTERNET OF (BATTLEFIELD) THINGS DEVICES USING DEEPEIGEN SPACE LEARNING

Abstract

In the military environment, the “Internet of Things (IoT)” usually contains a wide range of devices connected to the Internet and knots such as medical aids and wearable battle equipment. Cyber criminals, especially those supported by the government or nation, really want to get their hands on these devices and nodes IoT. Malware is a typical way of attack. This research introduces a deep approach based on learning to detect the “Internet Of Battlefield Things (IoBT) malware via the device’s Operational Code (OpCode)”. We transfer the OPCODES to vector space and use the learning method of deep Eigenspace to find out the difference between harmful and harmless applications. We also show that our proposed method for finding malware is strong and can build attacks that try to add unhealthy code. Finally, we patter our malware sample on Github to be used in a future study (for example, to test new malware detection methods).