Experimental Evaluation of Post-Quantum Homomorphic Encryption for Privacy-Preserving V2X Communication

2025-08-04

AI Summary

Privacy-preserving computation over sensitive traffic data in vehicular networks (V2X) demands efficient post-quantum homomorphic encryption (PQ-HE) schemes suitable for resource-constrained automotive environments. Method: This work presents the first end-to-end empirical evaluation of three PQ-HE schemes—BFV, BGV, and CKKS—in real vehicular settings, targeting two representative applications: encrypted traffic-flow counting and average-speed aggregation, deployed over Wi-Fi and Ethernet. All schemes are configured at 128-bit post-quantum security. Contribution/Results: BFV and BGV achieve end-to-end latency under 10 seconds, enabling real-time intersection monitoring; CKKS, though computationally heavier, remains viable for periodic numerical aggregation. The study establishes concrete performance boundaries for each scheme in intelligent transportation systems, providing the first empirical benchmark and practical deployment guidelines for privacy-enhancing V2X computing.

Technology Category

• Machine Learning: Quantum Machine Learning
• Application Domains: Transportation
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Security and Privacy: Privacy-enhancing technologies
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization

Abstract

Intelligent Transportation Systems (ITS) fundamentally rely on vehicle-generated data for applications such as congestion monitoring and route optimization, making the preservation of user privacy a critical challenge. Homomorphic Encryption (HE) offers a promising solution by enabling computation on encrypted data without revealing underlying content. This study presents the first real-world experimental evaluation of three post-quantum secure HE schemes, i.e., Brakerski-Fan-Vercauteren (BFV), Brakerski-Gentry-Vaikuntanathan (BGV), and Cheon-Kim-Kim-Song (CKKS), for vehicular communication scenarios. Two representative privacy-preserving use cases are considered: encrypted vehicle counting and average speed aggregation. Experiments are conducted over both Wi-Fi and Ethernet to assess performance under wireless and wired vehicle-to-everything (V2X) settings. Results show that BFV and BGV are suitable for latency-tolerant applications such as intersection monitoring and regional traffic analysis, with total end-to-end latencies under 10 seconds. While CKKS experiences higher overhead, it remains viable for periodic encrypted aggregation of numerical data. The experimental results demonstrate that HE can be feasibly deployed in ITS environments under 128-bit post-quantum security, provided that scheme-specific latency constraints are considered. This reinforces its potential to serve as a foundational tool for secure and privacy-preserving V2X data processing.

Problem

• Evaluates post-quantum homomorphic encryption for V2X privacy
• Compares BFV, BGV, CKKS schemes in real-world vehicular scenarios
• Assesses encrypted data aggregation for traffic monitoring applications

Innovation

• Evaluates post-quantum HE schemes for V2X
• Tests BFV, BGV, CKKS in real-world scenarios
• Demonstrates HE feasibility for ITS privacy