Enhancing Computational Efficiency and Time Optimization in Cloud IoT Intrusion Detection Using ANN and Hybrid Deep Learning

Balachandran, Swathy Menon (2024) Enhancing Computational Efficiency and Time Optimization in Cloud IoT Intrusion Detection Using ANN and Hybrid Deep Learning. Masters thesis, Dublin, National College of Ireland.

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

Abstract

Internet of Things (IoT) networks are essential across various domains in today's interconnected world including smart homes and industrial systems but they face significant security challenges due to their vulnerability to cyberattacks. This research presents a cloud-integrated intrusion detection system (IDS) specifically developed for IoT networks using advanced machine learning (ML) techniques to enhance security and efficiency. The system employs Artificial Neural Networks (ANN) for the rapid initial classification of normal and suspicious activities. Following this, a hybrid deep learning model that combines Convolutional Neural Networks (CNN) and Gated Recurrent Units (GRU) performs detailed analysis and classifies specific threats. This two-level approach optimizes resource usage by reserving computationally demanding processes for potentially malicious data. The architecture is developed to manage the substantial data volumes generated by IoT devices achieving 90.79% accuracy in initial classification and 74.32% in multiclass threat detection. It integrates with a Python flask application and deployed on Amazon Web Services (AWS) ensuring faster threat detection and response. Specifically, the AWS deployment reduced the threat detection time to approximately 0.33 seconds for normal traffic and around 1.92 seconds for attack files. This research highlights the significance of using advanced ML models that integrate with cloud-based solutions for enhance overall network resilience and reliability to address the unique security challenges in IoT environments. Future work required to validate the IDS using different datasets with regular updates to detect emerging threats effectively.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Gupta, Shaguna UNSPECIFIED
Uncontrolled Keywords:	Internet of Things; Intrusion Detection System; Cyber threats; Machine Learning; Hybrid Deep Learning
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Cloud computing 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 Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Cloud Computing
Depositing User:	Ciara O'Brien
Date Deposited:	03 Jul 2025 08:59
Last Modified:	03 Jul 2025 08:59
URI:	https://norma.ncirl.ie/id/eprint/8009

Actions (login required)

View Item