A Comprehensive Survey on IoT Attacks and Defenses: From Side Channels to AI-Driven Threats

The rapid proliferation of Internet of Things (IoT) devices has created an expansive and diverse attack surface that continues to challenge the security community. From smart home assistants and wearable devices to industrial control systems and autonomous vehicles, IoT ecosystems introduce unique vulnerabilities arising from constrained resources, heterogeneous communication protocols, and intimate physical-world interactions. This paper presents a comprehensive survey of IoT attack vectors and corresponding defense mechanisms, organized across four key dimensions: (1) side-channel attacks exploiting electromagnetic emissions, power consumption, and sensor data leakage; (2) network-level threats including botnet propagation, DNS manipulation, and traffic analysis; (3) adversarial machine learning attacks targeting IoT-integrated AI systems; and (4) emerging threats in augmented and virtual reality platforms. We further examine state-of-the-art defensive strategies spanning intrusion detection, authentication, firmware analysis, and AIassisted vulnerability discovery. By synthesizing findings from over 45 recent studies, we identify open challenges and outline future research directions for securing the next generation of IoT infrastructure.