Zero Trust-based Decentralized Identity Management System for Autonomous Vehicles

📅 2025-09-29
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
Traditional boundary-based security models and fragile identity management mechanisms fail to ensure trustworthiness for autonomous vehicles (AVs) operating in dynamic, untrusted V2X environments. To address this, this paper proposes the first lightweight, decentralized identity management framework for AVs that integrates zero-trust architecture with blockchain technology. Built on Hyperledger Iroha, the framework eliminates centralized authorities and enforces continuous, dynamic identity verification—adhering strictly to the “never trust, always verify” principle—while natively resisting spoofing and replay attacks. Experimental evaluation in an urban LTE-V2X setting demonstrates minimal system overhead: packet reception rate degradation remains below 7.5%, and channel utilization increases by less than 11%. The framework thus achieves a favorable trade-off among high security, low latency, and practical deployability, providing a scalable, trust-enabling identity infrastructure for cooperative AV operations.

Technology Category

Application Category

📝 Abstract
The rise of autonomous vehicles (AVs) promises to significantly enhance transportation safety and efficiency by mitigating human error, which is responsible for over 90% of road accidents. However, the increasing connectivity of AVs introduces new cybersecurity challenges, as traditional perimeter-based security models are inadequate for dynamic and untrusted environments. This paper presents a novel Zero Trust-based Decentralized Identity Management (D-IM) protocol for AVs. By integrating the core principles of Zero Trust Architecture, "never trust, always verify", with the tamper resistant and decentralized nature of a blockchain network, our framework eliminates reliance on centralized authorities and provides continuous verification for every entity. We detail the system's design, which leverages Hyperledger Iroha to enable lightweight and secure authentication without a central trusted entity. A comprehensive experimental evaluation, conducted across both urban and highway scenarios, validates the protocol's practicality. Our results demonstrate that the D-IM framework introduces minimal overhead, with less than 7.5% reduction in Packet Reception Rate (PRR) in urban settings and an increase of under 11% in Channel Busy Ratio (CBR) for LTE-V2X. These findings prove the protocol's efficiency and robustness, providing a resilient foundation for securing real-time V2X communication against impersonation and replay attacks.
Problem

Research questions and friction points this paper is trying to address.

Securing autonomous vehicle connectivity in dynamic environments
Replacing centralized authorities with decentralized identity management
Preventing impersonation and replay attacks in V2X communications
Innovation

Methods, ideas, or system contributions that make the work stand out.

Zero Trust-based decentralized identity management for autonomous vehicles
Leverages blockchain for tamper-resistant authentication without central authority
Uses Hyperledger Iroha for lightweight continuous verification in V2X communication
🔎 Similar Papers
No similar papers found.
A
Amal Yousseef
Department of Electrical and Computer Engineering, University of Arizona, Tucson, Arizona, USA
S
Shalaka Satam
Department of Electrical and Computer Engineering, University of Arizona, Tucson, Arizona, USA
Banafsheh Saber Latibari
Banafsheh Saber Latibari
Postdoctoral Research Associate, University of Arizona
Deep LearningSecurityEmbedded SystemsComputer Architecture
M
Mai Abdel-Malek
Department of Systems and Industrial Engineering, University of Arizona, Tucson, Arizona, USA
Soheil Salehi
Soheil Salehi
Assistant Professor, ECE, University of Arizona
IoT Hardware SecurityAI-enabled SecurityReconfigurable ComputingSpintronicsNeuromorphic Hardware
Pratik Satam
Pratik Satam
Assistant Professor, University of Arizona
Smart ManufacturingCyber SecurityMachine Learning