Edge-Based Information-Centric Networking for Enhanced Message Discrimination in Vehicular Ad-Hoc Networks

Abstract

Vehicular Ad Hoc Networks (VANETs) are central to Intelligent Transportation Systems (ITS), but practical deployment remains constrained by communication latency, scalability, and privacy requirements. This study develops an edge-based Information-Centric Networking (ICN) framework integrated with blockchain technology. Raspberry Pi units equipped with cameras perform real-time incident detection and classification using Convolutional Neural Networks (CNN) and You Only Look Once v8 (YOLOv8). Frame extraction, noise reduction, and data augmentation increased the reported detection accuracy from 70% to 97%. Severity-based incident prioritization reduced latency to 112 ms in low-density networks and 128 ms in high-density networks, while packet loss decreased from 12.12% to 6.34% and throughput increased from 600 to 700 Mbps. The ICN layer also enabled high-priority alerts to reach vehicle dashboards in less than 1 s. The evaluated configuration scaled to 400 mobile routers and 20 access points while maintaining low latency and stable communication. These findings indicate that the proposed architecture can support real-time, secure, and efficient vehicular communication in ITS environments.