The Role of Generative AI in Prediction, Data Processing, and Security for IoT Systems

The rapid expansion of the Internet of Things (IoT) has enabled unprecedented connectivity across industrial, urban, healthcare, and consumer domains. However, IoT networks face significant challenges in data heterogeneity, predictive analytics, and cybersecurity. Traditional methods often fall short in handling high-dimensional, incomplete, or dynamic data, as well as detecting evolving security threats. In recent years, Generative Artificial Intelligence (Generative AI), including Generative Adversarial Networks (GANS), Variational Autoencoders (VAEs), and Large Language Models (LLMs), has emerged as a promising solution to these challenges. This review provides a comprehensive analysis of Generative AI applications in IoT, focusing on three primary domains: prediction, data processing, and security. Key findings highlight the ability of generative models to forecast device behavior, optimize data handling, and enhance security through anomaly detection and attack simulation. The paper also identifies research gaps, including edge deployment constraints, privacy concerns, and the need for real-world validation, and outlines future directions such as Edge Intelligence, Federated Generative AI, and LLM-IoT integration. By synthesizing ۳۰ high-quality journal studies from ۲۰۲۲-۲۰۲۶, this review provides a unified perspective on how Generative AI can advance autonomous, adaptive, and resilient IoT networks.

کلیدواژه ها:

Internet of Things (IoT) ، Generative AI ، GANS ، Variational Autoencoders (VAEs) ، Predictive Analytics ، IoT Security