A Review: State-of-the-Art of Integrating AI Models with Moving-target Defense for Enhancing IoT Networks Security

Abstract

The rapid proliferation of IoT devices across various sectors has significantly expanded the attack surface for cyber threats. Traditional security strategies, which rely on static defenses, often fail to adequately protect these diverse and resource-constrained devices, rendering them vulnerable to attacks. To address this challenge, proactive security mechanisms like Moving-Target Defense (MtD) introduce dynamic, real-time changes, making systems less predictable and more challenging for attackers to exploit. However, deploying MtD strategies in IoT networks presents various limitations and trade-offs between the network’s trustworthiness, performance, and compatibility. To overcome these challenges, AI-based MtD strategies offer a promising solution by dynamically adapting and optimizing defense mechanisms, thereby enhancing system resilience and unpredictability. This paper systematically reviews the use of Artificial Intelligence (AI) models to improve security in IoT networks, with a particular focus on integrating AI with MtD techniques to bolster network resilience. The review adheres to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) model principles and employs a systematic mapping process to analyze AI’s effectiveness in predicting and mitigating IoT attacks while exploring the potential of MtD mechanisms.