A Scalable Cluster Position-Based Routing Protocol for VANET

Abstract

Due to the rapid development of wireless communication technology and the growing demand for services, it is expected that the emergence of Vehicular Ad hoc Networks (VANET)s would enable a variety of applications such as driver assistance, traffic efficiency, and road safety. Frequent changes that occur in the network often leads to major challenges in VANET, such as dynamic topology changes, shortest routing paths and also scalability due to the high dynamic topology where the number of vehicles on the road increases and decreases rapidly. One of the best solutions for such challenges is to divide the network into clusters and then choose a Cluster Head (CH) in each cluster to ensure appropriate message transmission in the VANET. In order to resolve the network scalability issue and accommodate additional applications in VANETs, efficient clustering methods are suggested. Because only the CH communicates with the Road Side Units (RSU) and delivers relevant messages, there is a potential reduction in the communication overhead between RSUs and other VANET components. However, clustering algorithms are necessary to ensure the stability of the cluster because of the dynamic nature of VANETs' network topology. The selection of CH is a crucial step in the clustering process. In order to improve cluster stability and data transmission efficiency, this thesis proposes a Clustering technique based on Ad hoc On-Demand Distance Vector routing protocol (AODVC) for VANETs that is cluster position-based. It also implements Vehicle-to-Vehicle (V2V) and CH-to-RSU communication. We also provided an approach for choosing a suitable vehicle to serve as the CH. The distance the vehicle is from the cluster boundary and its average speed are considered when choosing the CH. Our proposed protocol is implemented using the OMNeT++ 5.5.1 simulator, the simulation results show that the suggested technique enhances some important metrics such as packet delivery fraction, the number of control packets transmitted for each delivered data packets, average path length, and average route latency when compared with the standard AODV in terms of node mobility speed, node density, number of clusters, and network sizes. Where the simulation results show that the AODV-C protocol outperforms AODV in terms of reliability by around 12% for increasing vehicle speed, 23% for increasing vehicle number, 10% for increasing network size and local traffic, and 20% for increasing cluster number. In addition, improved latency by 10% for increasing vehicle speed, 13% for increasing vehicle number, 10% for expanding network size, 11% for increasing local traffic, and 12% for increasing cluster number. The proposed protocol guarantees scalability by having a good packet delivery fraction and low control packets transmitted for each delivered data packets even when the network size increase and the number of vehicles along with their speeds increase too. XIV Keywords: Intelligent Transportation Systems (ITS), Vehicular Ad-hoc Networks (VANETs), clustering.