Rogue Node Detection Based on a Fog Network Utilizing Parked Vehicles

• Published in: Applied sciences Vol. 13; no. 2; p. 695
• Main Authors: Hua, Jiwei, Zhang, Bo, Wang, Jinao, Shao, Xin, Zhu, Jinqi
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.01.2023
• Subjects: Accuracy; Algorithms; False information; Fog; fog networks; Hypothesis testing; Internet of Vehicles; Intrusion detection systems; Machine learning; Nodes; parked vehicles; Roads & highways; rogue nodes; Support vector machines; Traffic congestion; Urban areas; Vehicles; Wireless communications
• ISSN: 2076-3417; 2076-3417
• DOI: 10.3390/app13020695

Rogue nodes in the Internet of vehicles (IoV) bring traffic congestion, vehicle collision accidents and other problems, which will cause great social losses. Therefore, rogue node discovery plays an important role in building secure IoV environments. Existing machine learning-based rogue node detection methods rely too much on historical data, and these methods may lead to long network delay and slow detection speed. Moreover, methods based on Roadside Units (RSUs) have poor performance if the number of RSUs is insufficient. Based on the widespread presence of ground vehicles, we propose a rogue node detection scheme based on the fog network formed by roadside parked vehicles. To achieve efficient rogue node discovery, a fog network composed of stable roadside parked vehicles is dynamically established, in which each fog node firstly collects the information of moving vehicles on the road in its coverage range, and then fog nodes use the U-test method to determine the rogue nodes in parallel, so as to find the bad nodes efficiently. Simulation results show that the proposed algorithm has higher detection accuracy and stability than the other rogue node detection schemes.