Cross-Layer Optimization for Enhancing QoS in Software-Defined Vehicular Networks

The current study offers a cross-layer optimization for improving the QoS in Software Defined Vehicular Networks or SDVNs for short. SDVNs enhance the control of vehicular communication using the features of SDN, but managing different quality of services is still a critical issue under the dynamic architecture. Most of the conventional QoS optimization strategies optimize the QoS in a single layer resulting in inefficiency while handling the variability in traffic that may exist in vehicular networks. In response to this, we introduce the cross layer optimization methodology to come up with a mechanism that would enable comprehensive control of resource use, routing, and traffic management at the different layers within a network. The proposed framework is developed and simulated with NS-3 and SUMO simulators where TAPAS Cologne and LuST vehicular mobility datasets are utilized for the evaluation. Associate heuristic algorithms such as, Particle Swarm Optimization (PSO) and Genetic Algorithms (GA) are used for dynamic resource allocation and optimal routing choices. The simulation results also reveal that the proposed system yields enhanced latency, throughput, packet delivery ratio, and resource use effectiveness than the previous QoS optimization techniques. This research presents a feasible and flexible solution to the QoS issues in SDVNs and thus contributes to the development of ITS.