Edge Computing for Intelligent Internet of Things: Architectures, Applications, Challenges, and Emerging Trends

Abstract

The rapid growth of the Internet of Things (IoT) has intensified the demand for computing paradigms capable of supporting real-time data processing, intelligent decision-making, and scalable service delivery. Edge computing has emerged as a key solution by relocating computation and analytics closer to data sources, thereby reducing latency, bandwidth consumption, and dependence on centralized cloud infrastructures. This review provides a comprehensive synthesis of recent advances in edge computing for intelligent IoT, focusing on its foundational concepts, architectural frameworks, intelligent computing techniques, application domains, implementation challenges, and emerging technological trends. The review examines the integration of artificial intelligence, machine learning, federated learning, TinyML, cloud–edge–fog collaboration, and next-generation communication technologies that collectively enhance the efficiency and adaptability of distributed IoT ecosystems. It further discusses the growing adoption of edge computing in smart manufacturing, healthcare, transportation, agriculture, and smart infrastructure, highlighting its contribution to secure, responsive, and data-driven operations. The review also identifies critical challenges, including security, privacy, interoperability, energy efficiency, and regulatory compliance, that continue to influence large-scale deployment. Finally, it outlines emerging research directions involving Generative AI, Digital Twin Networks, 6G communication, open virtualization, and autonomic computing as promising enablers of future intelligent edge environments. Overall, this review offers a consolidated perspective on the evolving landscape of edge computing and identifies key research opportunities for developing scalable, resilient, and sustainable intelligent IoT systems.