Enhancing Cybersecurity in IoT Healthcare Systems: A CNN-GRU Hybrid Approach for Intrusion Detection

Usman, Ashna (2024) Enhancing Cybersecurity in IoT Healthcare Systems: A CNN-GRU Hybrid Approach for Intrusion Detection. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (706kB) | Preview

Abstract

The proliferation of IoT devices has revolutionized industries such as healthcare, smart homes, and manufacturing, but has also made them prime targets for sophisticated intrusion attacks. Traditional intrusion detection methods become inefficient in IoT networks because of the dynamic nature of IoT and due to issues like model shift, problem of generalization and ineffectiveness when dealing with zero-day attacks, etc. To overcome these drawbacks, this research adopts a new approach that involves Convolutional Neural Networks (CNN) and Gated Recurrent Units (GRU). It is therefore important to use simple mathematical equations, so that the model that results from this research can be implemented in real-time; preferably in low-powered IoT devices. Using the IoT Healthcare Security Dataset sourced from Kaggle, data preprocessing, normalization, and features extraction were conducted to improve the efficiency of the model proposed. CNN-GRU is computationally efficient and has high detection accuracy and the model can be deployed in real-time in resource-constrained IoT devices. The efficiency of the model is specifically described by 99.52% accuracy, 98.7% precision, 99.6% recall, and 99.49 F1 score of detection. These results place the model to be used as a reference against which future IoT security frameworks become benchmark-able based on the effective approach towards complex IoT intrusion detection that is both scalable and resource-efficient. The proposed CNN-GRU approach outperforms CNN-LSTM (2021), RLSTM (2022) in terms of performance metrics. Despite these benefits it is crucial to perform further validation on different IoT environments and introduce other types of threats to reduce threats to generalizability and make it more flexible

Item Type:	Thesis (Masters)
Supervisors:	Name Email Heffernan, Niall 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 R Medicine > Healthcare Industry 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 Cyber Security
Depositing User:	Ciara O'Brien
Date Deposited:	28 Jul 2025 14:54
Last Modified:	28 Jul 2025 14:54
URI:	https://norma.ncirl.ie/id/eprint/8273

Actions (login required)

View Item