A Cross-Layer Review of Intelligent, Secure, and Privacy-Preserving Internet of Vehicles

A Cross-Layer Review of Intelligent, Secure, and Privacy-Preserving Internet of Vehicles (Alayedi & Jaradat, 2026) provides a unified framework analyzing the interactions among AI/ML/DL models, blockchain-enabled trust, and cybersecurity defenses across five architectural layers: perception, communication, edge/cloud computing, blockchain-enabled trust, and application services.

The survey identifies key future research directions including lightweight blockchain consensus, explainable AI (XAI) for safety-critical decisions, federated intrusion detection under non-IID data, semantic interoperability, and digital twin-assisted security evaluation. It positions its contribution by comparing prior literature, noting that existing surveys often focus narrowly on specific technologies like trust management, edge optimization, or blockchain, lacking a comprehensive synthesis of cross-layer trade-offs between latency, scalability, privacy, and security.

Key insights include: Cross-Layer Framework: The paper establishes a structured approach to understanding how intelligence, security, and privacy mechanisms interact, addressing gaps in isolated technology analyses. Threat-Defense Mapping: It details specific threats such as spoofing, Sybil attacks, and data poisoning, mapping them to intelligent detection models and cryptographic mechanisms across different IoV layers. Privacy-Preserving Intelligence: The review evaluates techniques like Federated Learning (FL), differential privacy, and homomorphic encryption, highlighting the balance between privacy preservation, model accuracy, and computational overhead. Limitations of Prior Work: Previous surveys are critiqued for limited coverage of interoperability, blockchain integration, or privacy-preserving computation, whereas this work offers a holistic view of design trade-offs and a roadmap for scalable, trustworthy IoV systems.

Scope and focus. The paper presents a cross-layer review of security, intelligence, and privacy mechanisms for the Internet of Vehicles (IoV), with emphasis on techniques that are practical in latency-sensitive, resource-constrained vehicular environments. It examines how lightweight blockchain, explainable AI (XAI), federated intrusion detection systems (IDS), and digital-twin-assisted security can be integrated across vehicle, edge, and network layers to support intelligent and trustworthy IoV operations.

Key contributions and insights. A central contribution is the cross-layer framing: rather than treating blockchain, AI, intrusion detection, and digital twins as isolated technologies, the review considers how they interact to address trust, interpretability, collaborative defense, and predictive security. Lightweight blockchain is positioned as a candidate for decentralized identity, transaction logging, and tamper-resistant coordination without the overhead of full-scale distributed ledgers. XAI is treated as essential for making AI-driven vehicle decisions auditable and operator-trustworthy, while federated IDS enables distributed threat detection without exposing raw vehicle or user data. Digital twins are highlighted as a means to model, simulate, and monitor IoV behavior for anomaly detection and security validation. The paper also compares prior surveys, helping to clarify where earlier work has focused and what gaps remain.

Why it matters. The work is relevant because IoV systems combine safety-critical vehicle control, V2X communication, and large-scale data exchange, making security and privacy failures potentially consequential. By synthesizing these approaches into a coherent cross-layer perspective, the review offers a useful reference for researchers and system designers seeking to balance real-time performance, privacy preservation, explainability, and distributed trust in intelligent vehicular networks.