A Survey on Cross-Architectural IoT Malware Threat Hunting

Anandharaju Durai Raju,Ibrahim Y Abualhaol,Ronnie Salvador Giagone,Shengqiang Huang,Yang Zhou

doi:10.1109/access.2021.3091427

Anandharaju Durai Raju, Ibrahim Y Abualhaol + Show 3 more

Open Access

https://doi.org/10.1109/access.2021.3091427

Copy DOI

Abstract

In recent years, the increase in non-Windows malware threats had turned the focus of the cybersecurity community. Research works on hunting Windows PE-based malwares are maturing, whereas the developments on Linux malware threat hunting are relatively scarce. With the advent of the Internet of Things (IoT) era, smart devices that are getting integrated into human life have become a hackers' highway for their malicious activities. The IoT devices employ various Unix-based architectures that follow ELF (Executable and Linkable Format) as their standard binary file specification. This study aims at providing a comprehensive survey on the latest developments in cross-architectural IoT malware detection and classification approaches. Aided by a modern taxonomy, we discuss the feature representations, feature extraction techniques, and machine learning models employed in the surveyed works. We further provide more insights on the practical challenges involved in cross-architectural IoT malware threat hunting and discuss various avenues to instill potential future research.

Highlights

E ACH day the digital world is exposed to millions of new malware (Malicious Software) attacks, and almost all of them are oblivious to the day-to-day users while they happen
TAXONOMY we provide a taxonomy of feature representations used for static analysis-based malware threat hunting in the Internet of Things (IoT) landscape and highlight the features useful for cross-architectural IoT malware threat hunting that requires the abilities of instruction set architecture (ISA) neutrality and operating systems (OS) platform independence as described in Sections II-B and II-C respectively
We present the existing survey works by discussing general Linux surveys and IoT surveys first, followed by Crossarchitecture specific IoT surveys

Summary

Introduction

E ACH day the digital world is exposed to millions of new malware (Malicious Software) attacks, and almost all of them are oblivious to the day-to-day users while they happen. In the past two decades, the machine learning approaches adapted to the domain of malware detection/classification strove towards convergence at better handling of malware threats as hard as zero-day attacks. Deep learning approaches are taking part in the arena to cope with the explosion of malware variants. While some financially motivated attack groups focus on big industry players, surprisingly, more than 60% of the attacks are directed towards small and mid-sized businesses [3]. Such cybercrimes are estimated to cause global damage of 6 Trillion in 2021 [3] and are expected to reach 10.5 Trillion by 2025.

Objectives

Findings

Discussion

Conclusion