Artificial Intelligence Techniques for Malicious Node Detection with Cross-Attention Vision Transformers and Blockchain-Based Distributed Data Storage in Wireless Sensor Networks

Abstract

Wireless Sensor Networks (WSNs) are increasingly deployed in critical domains such as environmental monitoring, healthcare, and industrial automation. However, their decentralized and resource-constrained nature makes them highly vulnerable to malicious node attacks, including data manipulation, packet dropping, and routing disruption. Traditional detection mechanisms based on rule-based or trust-based approaches are insufficient to handle sophisticated and dynamic attack patterns. Recent advancements in Artificial Intelligence (AI), particularly deep learning and transformer-based architectures, have significantly enhanced detection capabilities. Vision Transformers (ViTs), equipped with cross-attention mechanisms, enable global feature extraction and multi-node correlation, improving anomaly detection accuracy. Simultaneously, blockchain technology provides a decentralized and tamper-proof framework for secure data storage, node authentication, and trust management, eliminating single points of failure. This paper presents a comprehensive review of AI-driven malicious node detection techniques integrating cross-attention Vision Transformers and blockchain-based distributed storage in WSNs. It analyzes research developments, focusing on hybrid AI-blockchain frameworks, graph-based learning, and federated learning approaches. A comparative analysis is conducted based on accuracy, scalability, energy efficiency, and computational complexity. The study also identifies key challenges such as resource constraints, energy consumption, and integration complexity. Finally, future research directions are discussed to enable secure, scalable, and intelligent WSN infrastructures.