Explanatory study on the role of neural networks in maintaining cyber security in IoT

Mathew, Albin (2024) Explanatory study on the role of neural networks in maintaining cyber security in IoT. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (612kB) | Preview
Preview	PDF (Configuration Manual) Download (663kB) | Preview

Abstract

The constantly rising use of IoT devices has created a necessity for strong detection and response models to identify anomalies, intrusion, and threats within vast networks. In this work, the authors present and extensive system for using both real and synthetic datasets for multi-task anomaly and intrusion detection in IoT systems. It integrates state-of-the-art predictive feature extraction and other pre-processing steps, alongside a modularity-based deep learning architecture for models adapted to each application area. The framework uses specific Feed-Forward Neural Networks (FFNNs) to address varied situations such as use and device authentication and prediction of future threats. Clearly, real-world IoT traffic is augmented by synthetic datasets created with the help of the Faker library to simulate as real-life devices and possible attacks. Each dataset is scaled, split into train and test sets, and trained in an independent manner. The FFNNs are intensified with dropout layers to modify and scale off-centre neurons, batch normalisation to scale and regularise low and high neurons, and the Adam optimiser to optimise the neurons to optimum outcomes. Real-time anomaly detection is also enabled by the framework because models built with the framework can predict task-specific outcomes in real-time. Based on evaluation metrics and performance visualisation, it is proved that the system can effectively distinguish normal and suspicious behaviours for further practical applications in IoT scenarios.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Pantridge, Michael 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 Cyber Security
Depositing User:	Ciara O'Brien
Date Deposited:	23 Jul 2025 15:03
Last Modified:	23 Jul 2025 15:03
URI:	https://norma.ncirl.ie/id/eprint/8224

Actions (login required)

View Item