Artificial Intelligence Techniques for Trusted Cloud-Enabled IoT Networks Using Blockchain and Siamese Heterogeneous Convolutional Neural Networks: Trends and Challenges

Abstract

The rapid proliferation of Internet of Things (IoT) devices and cloud computing infrastructures has introduced significant challenges related to security, trust management, and data integrity. Traditional centralized architectures are increasingly inadequate for handling dynamic, large-scale, and heterogeneous IoT environments. This paper explores advanced artificial intelligence (AI) techniques for building trusted cloud-enabled IoT networks by integrating blockchain technology and Siamese heterogeneous convolutional neural networks (SHCNNs). Blockchain provides decentralized trust, immutability, and transparency, while Siamese architectures enable similarity learning for anomaly detection and malicious node identification. Recent research highlights the importance of combining AI and blockchain to enhance trust management, data privacy, and system resilience in distributed IoT systems. This study presents a comprehensive review of recent developments in AI-driven trust mechanisms, focusing on deep learning models, blockchain-based frameworks, and hybrid cloud-edge architectures. A comparative analysis of existing approaches is conducted to evaluate performance metrics such as accuracy, scalability, and computational efficiency. Furthermore, key challenges including data heterogeneity, scalability limitations, and explainability issues are discussed.

The paper concludes by outlining future research directions involving federated learning, edge intelligence, and explainable AI for secure IoT ecosystems.