Internet of Things (IoT) in military environment for the most part comprises of a different scope of Internet-associated gadgets and hubs (for example clinical gadgets to wearable battle outfits), which are an important objective for digital crooks, especially state-supported or country state on-screen characters. A typical assault vector is the utilization of malware. In this paper, we present a profound learning- based strategy to recognize the Internet of Battlefield Things (IoBT) malware by means of the gadget’s Operational Code (OpCode) arrangement. We transmute OpCodes into a vector space and apply a profound Eigen space learning way to deal with group malevolent and benign application. We additionally exhibit the strength of our proposed approach in malware identification and its manageability against garbage code inclusion assaults. Finally, we make accessible our malware test on GitHub which ideally will profit future exploration endeavors (for example for assessment of proposed malware recognition draws near).