I. Introduction
With the rapid development of the Internet of Vehicles (IoV), the IoV has become an essential component of Intelligent Transportation Systems (ITS) [1]. Vehicles can access a wide range of delay-sensitive and location-aware services, for example, road traffic monitoring [2], emergency incident reporting [3], and cloud-assisted autonomous driving [4] to improve traffic conditions and driving safety. As reported in [5], it is possible to improve road traffic congestion by 60% and short distance transportation by 70% using IoV. The IoV enhances the user experience by providing flexibility, scalability, seamless connectivity, and less power consumption. However, in the IoV, where the system communicates over the public network, it is vulnerable to various attacks such as man-in-the-middle attacks, replay attacks, and impersonation attacks [6]. Moreover, the information about identity and location is delivered along with messages, and leakage of these data can create privacy issues. Generally, anonymous authentication is considered a viable solution to protect the private data of vehicles [7]. Such as standards IEEE 1609.2 [8] in the United States and ETSI [9] in Europe indicate that the utilization of Public Key Infrastructure (PKI) certificate systems to protect vehicular communications; [10], [11] proposed certificateless authentication scheme for IoV based on the assistance of trusted authority. However, the rapid development of ITS has led to an explosion in the number of vehicles and diverse demands for cloud services. It is a challenge to ensure that vehicular entities enjoy cloud services from different cloud service providers (CSPs) in a time-efficient and reliable manner [12]. The traditional anonymous authentication schemes, which have not considered entities needing to switch and connect efficiently between multiple CSPs, will no longer be applicable. Researchers have extended the study of single cloud services to multiple cloud services [13]. Meanwhile, the heterogeneity, high mobility, and low latency of IoV will pose significant challenges. Hence, the IoV in multi-server scenarios should be distributed, decentralized, scalable, and flexible to adapt to the future development of IoV and fully exploit the potential of ITS [14].