Security and Trust Management in the Internet of Vehicles (IoV): Challenges and Machine Learning Solutions

Abstract

The Internet of Vehicles (IoV) is a technology that is connected to the public internet and is a subnetwork of the Internet of Things (IoT) in which vehicles with sensors are connected to a mobile and wireless network. Numerous vehicles, users, things, and networks allow nodes to communicate information with their surroundings via various communication channels. IoV aims to enhance the comfort of driving, improve energy management, secure data transmission, and prevent road accidents. Despite IoV’s advantages, it comes with its own set of challenges, particularly in the highly important aspects of security and trust. Trust management is one of the potential security mechanisms aimed at increasing reliability in IoV environments. Protecting IoV environments from diverse attacks poses significant challenges, prompting researchers to explore various technologies for security solutions and trust evaluation methods. Traditional approaches have been employed, but innovative solutions are imperative. Amid these challenges, machine learning (ML) has emerged as a potent solution, leveraging its remarkable advancements to effectively address IoV’s security and trust concerns. ML can potentially be utilized as a powerful technology to address security and trust issues in IoV environments. In this survey, we delve into an overview of IoV and trust management, discussing security requirements, challenges, and attacks. Additionally, we introduce a classification scheme for ML techniques and survey ML-based security and trust management schemes. This research provides an overview for understanding IoV and the potential of ML in improving its security framework. Additionally, it provides insights into the future of trust and security enhancement.

An Effective Ensemble Learning-Based Real-Time Intrusion Detection Scheme for an In-Vehicle Network

“…Attackers can utilize these vulnerabilities to compromise the communication system's availability, confidentiality, and integrity [38][39][40]. In [10], hacking was employed to gain unauthorized access and control over the functionalities of a Jeep Cherokee.…”

Artificial intelligence powered internet of vehicles: securing connected vehicles in 6G

“…The primary objective of this algorithm is to identify deviations from normal traffic patterns in V2X communications. Machine learning algorithms, particularly those employing unsupervised learning, analyze historical traffic data to learn normal patterns [41], [42]. These models can detect anomalies such as sudden traffic increases or unusual message volumes, allowing for timely alerts to potential attacks.…”

“…The objective of this algorithm is to classify V2X messages as either benign or malicious based on their content, behavior, and contextual information. Deep learning models, often utilizing neural networks, are capable of processing message content and behavior to learn normal patterns and differentiate them from malicious ones [41]- [44]. This real-time intrusion detection capability is vital for preventing unauthorized access and potential threats.…”

Navigating the Landscape of Autonomous Buses: Insights in Ibaraki, Japan

“…Chassis execution, integrating actuators and control systems, ensures precise maneuvers, contributing to passenger comfort and confidence [27]. The integration of IoV facilitates real-time data exchange and communication between vehicles and infrastructure, optimizing traffic and enhancing safety features [28]. These technologies collectively revolutionize public transportation, offering efficient, safe, and sustainable mobility solutions for urban environments and shaping the future of autonomous transportation ecosystems.…”