A Comprehensive Review of Secure Threat Detection Models and Emerging Paradigms for IoT Networks

Abstract

The rapid proliferation of the Internet of Things (IoT) has introduced a vast and complex attack surface, intensified by the resource-constrained nature of devices and a lack of universal security standards. This review provides a comprehensive analysis of secure threat detection models and emerging paradigms designed to address these challenges. The article first establishes a systematic taxonomy of cybersecurity threats across the IoT's architectural layers such as Perception, Network, and Application. It then delves into a critical examination of advanced intrusion detection systems (IDS), with a focus on models leveraging machine learning (ML), deep learning (DL), and hybrid architectures. The performance of these models is evaluated against key metrics and benchmark datasets. The review further explores cutting-edge concepts such as federated learning for privacy-preserving detection, adversarial machine learning, and post-quantum cryptography. This synthesis offers a strategic overview of the current landscape, identifying research gaps and outlining a path toward building resilient, adaptable, and trustworthy IoT ecosystems for the future.