AI-Driven Secure Intrusion Detection for Internet of Things (IOT) Networks
Chigozie K Ejeofobiri *
School of Computer Science and Digital Technologies, University of East London, London, UK.
Olayinka Olubola Victor-Igun
Department of Computer Engineering, University of Benin, Benin, Nigeria.
Clifford Okoye
Department of Biochemistry and Molecular Biology, The University of Georgia, Athens, USA.
*Author to whom correspondence should be addressed.
Abstract
The rapid proliferation of Internet of Things (IoT) devices has transformed various sectors, enhancing connectivity and efficiency. However, this surge has also introduced significant security vulnerabilities, making IoT networks attractive targets for cyber threats. This literature review investigates the development of AI-powered intrusion detection systems (IDS) tailored specifically for IoT environments. By leveraging machine learning algorithms, these systems can analyze vast amounts of data generated by IoT devices, identifying anomalous patterns indicative of potential security breaches. The review categorizes existing machine learning techniques, including supervised, unsupervised, and reinforcement learning approaches, assessing their effectiveness in real-time anomaly detection and response. Furthermore, the Key challenges, including computational and energy constraints, are discussed, alongside advanced approaches like feature selection and hybrid models to enhance detection accuracy with minimal resources. Ultimately, this review highlights the necessity for a multi-layered security framework that not only addresses current threats but also anticipates future challenges posed by evolving cyberattack methodologies. By synthesizing insights from recent studies, the findings aim to inform the design of more robust and adaptive AI-powered IDS, contributing to the secure implementation of IoT networks across diverse applications.
Keywords: Internet of things, intrusion detection system, machine learning, cybersecurity, anomaly detection