Detecting Adversarial Network Behaviors in IoT Environment

Bafna, Tejas Sandeep (2024) Detecting Adversarial Network Behaviors in IoT Environment. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (7MB) | Preview

Abstract

The Internet of Things (IoT) has expanded rapidly across industries and enterprises bringing innovation and value to various sectors, while at the same time exposing critical cybersecurity risks arising from the growing complexity, heterogeneity, and resource constraints of IoT systems. IDS are not able to deal with threats such as zero-day threats, insider threat, encrypted traffic, polymorphic viruses and traffic camouflage. They also experience issues with low-and-slow attacks, IoT exploits, and Advanced Persistent Threats (APTs) that are stuck in normal behavior patterns, thus leaving the opportunity for detection gaps and false negatives. In the context of this work, deep learning models for IoT intrusion detection are examined, with a special emphasis on CNNs and the proposed Conv-GAN model for data augmentation. The standalone CNN model was tested with the RT-IoT 2022 dataset and showed excellent performance with 99.3% accuracy and good detection of most of the attacks. The Conv-GAN part of the feature set, when combined with synthetic data to tackle class imbalance, showed difficulties in synthetic data quality and incorporation leading to decreased performance compared with CNN. The results presented in this paper confirm the ability of CNNs and prove their potential for IoT intrusion detection, as well as identify further development possibilities for hybrid models.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Clifford, William UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QA Mathematics > Computer software > Computer Security T Technology > T Technology (General) > Information Technology > Computer software > Computer Security T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Telecommunications > Computer networks > Internet of things
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	01 Sep 2025 14:57
Last Modified:	01 Sep 2025 14:57
URI:	https://norma.ncirl.ie/id/eprint/8679

Actions (login required)

View Item