A New Cryptojacking Malware Classifier Model Based on Dendritic Cell Algorithm

Abstract

A new threat known as "cryptojacking" has entered the picture where cryptojacking malware is the future trend for cyber criminals, who infect victim's device, install cryptojacking malware, and use the stolen resources for crytocurrency mining. Worse comes to worst by 2020, researchers estimate there will be 30 billion of IoT devices in the world. IoT devices are vulnerable to attacks because of their simple configuration, unpatched vulnerability and weak passwords. IoT devices also prone to be poorly monitored because of their nature. There is lack of studies that provide in depth analysis on cryptojacking malware classification using machine learning approach where the current research mostly focused on manual analysis of web-based cryptojacking attacks. As IoT devices requires small processing capability, a lightweight model are required for the cryptojacking malware detection algorithm to maintain its accuracy without sacrificing the performance of other process. As a solution, we propose a new lightweight cryptojacking classifier model based on machine learning technique that may fit in a low processing capability environment such as IoT and Cyber Physical Systems (CPS). This paper aim to disscuss a new approach based on dendritic cell algorithm in order to provide a lightweight cryptojacking classifier model. The output of this paper will be significant used in detecting cryptojacking malware attacks that benefits multiple industries including cyber security contractors, oil and gas, water, power and energy industries.