Over time, wireless technology is evolving impulsively. For the most part, studies presently focused on the telecommunication field. Vehicular Ad-hoc Network (VANET) is one of the leading genuine and appealing areas for Intelligent Transportation Systems (ITS) originators and organizations. Essentially, VANET is an ad-hoc network consisting of several vehicles that flow on the same road. The existence of such networks revolutionizes the wireless communication field. The immense increase in the number of smart vehicles throughout time is the reason behind the massive evolution in VANET technologies. VANETs requirements are evolving quickly, therefore, having more flexible, scalable, and well-connected routes in VANETs is becoming essential in any new routing protocol. Antiquated VANETs routing protocols and designs encounter very challenging technical, employment, and control problems with the evolution of VANETs requirements. This paper presents a new efficient routing protocol for VANETs, namely, Innovative Cluster-Based Dual-Phase Routing Protocol Using Fog Computing and Software-Defined Vehicular Network (ICDRP-F-SDVN) which utilizes state-of-the-art technologies to overcome the drawbacks of traditional VANETs routing protocols. The use of fog computing and Software-Defined Networks (SDN) combined in our protocol presents a robust new architecture that meets all the new requirements and overcomes challenges arising from technological development and rapid escalation in the number of smart vehicles. Additionally, we propose a novel clustering algorithm mainly for the reason of lessening the long-distance communications in each cluster. In conjunction with this clustering algorithm, we propose an efficient control overhead reduction mechanism, thereby lowering the messages' exchanges imposed. Interestingly, our proposed protocol is a dual-phase protocol with a 90% of SDN packet delivery ratio. In other words, when the SDN fails to deliver packets, traditional Ad-hoc On-Demand Distance Vector (AODV) handles them as a fallback mechanism. In this paper, the performance of ICDRP-F-SDVN is evaluated in terms of throughput and end-to-end delay and further compared with four closely related routing protocols, that are, Intersection Dynamic VANET Routing (IDVR), Vehicle Density and Load Aware (VDLA), Intelligent Routing Protocol Using Real-Time Traffic Information in Urban Vehicular Environment (IRTIV), and Greedy Perimeter Coordinator Routing (GPCR). In terms of control overhead, the comparison is conducted with Control Overhead Reduction Algorithm (CORA), Moving Zone-Based Routing Protocol (MoZo), Beaconless Routing Algorithm for Vehicular Environment (BRAVE), and Cluster-Based Directional Routing Protocol (CBDRP). Our proposed protocol has achieved impressive simulation results compared to previously mentioned protocols. In particular, the throughput is increased by 8277%, 64309%, 76235%, and 96053% while the end-to-end delay is reduced by 77.5%, 82.5%, 87.2%, 95.6% over IDVR, VDLA, IRTIV, and GPCR, respectively. Finally, the control overhead is lowered by 42.3%, 87%, 93.5%, 96.7% over that of CORA, MoZo, BRAVE, and CBDRP, respectively.

随着时间的推移，无线技术正在迅速发展。在大多数情况下，目前的研究集中在电信领域。车载 Ad-hoc 网络 (VANET) 是智能交通系统 (ITS) 发起者和组织的主要真正和吸引人的领域之一。本质上，VANET 是一个 ad-hoc 网络，由在同一条道路上行驶的几辆汽车组成。这种网络的存在彻底改变了无线通信领域。随着时间的推移，智能汽车数量的巨大增长是 VANET 技术大规模发展的原因。VANET 的要求正在迅速发展，因此，在 VANET 中拥有更灵活、可扩展和连接良好的路由对于任何新的路由协议都变得至关重要。随着 VANET 要求的发展，过时的 VANET 路由协议和设计遇到了非常具有挑战性的技术、使用和控制问题。本文提出了一种新的高效 VANET 路由协议，即创新的基于集群的基于雾计算和软件定义车辆网络的双相路由协议 (ICDRP-F-SDVN)，它利用最先进的技术来克服传统VANETs路由协议的缺点。在我们的协议中结合使用雾计算和软件定义网络 (SDN) 提供了一个强大的新架构，可以满足所有新要求并克服技术发展和智能汽车数量快速增长带来的挑战。此外，我们提出了一种新的聚类算法，主要是为了减少每个聚类中的长距离通信。结合这种聚类算法，我们提出了一种有效的控制开销减少机制，从而降低了强加的消息交换。有趣的是，我们提出的协议是一个双阶段协议，具有 90% 的 SDN 数据包传递率。换句话说，当 SDN 无法传递数据包时，传统的 Ad-hoc 按需距离矢量 (AODV) 将它们作为一种回退机制来处理。在本文中，ICDRP-F-SDVN 的性能在吞吐量和端到端延迟方面进行了评估，并进一步与四种密切相关的路由协议进行了比较，即交叉口动态 VANET 路由（IDVR）、车辆密度和负载意识到（VDLA），在城市车辆环境中使用实时交通信息的智能路由协议 (IRTIV) 和贪婪周界协调路由 (GPCR)。在控制开销方面，分别与控制开销减少算法（CORA）、基于移动区域的路由协议（MoZo）、车载环境无信标路由算法（BRAVE）和基于集群的定向路由协议（CBDRP）进行了比较。与前面提到的协议相比，我们提出的协议取得了令人印象深刻的模拟结果。尤其是吞吐量提升了8277%、64309%、76235%、96053%，而端到端延迟比IDVR、VDLA、IRTIV、IRTIV降低77.5%、82.5%、87.2%、95.6% GPCR，分别。最后，控制开销比 CORA、MoZo、BRAVE 和 CBDRP 分别降低了 42.3%、87%、93.5%、96.7%。