Software Defined Network based Fog Architecture for Workload Balancing in Autonomous Vehicles

Cloud technology has been one of the important approaches in Autonomous Vehicles (AV), which have been central to many studies in today's history. While the technology offers high computation, memory and network connectivity performance, the AV has always had little flexibility and scalability, poor location-based services and large process delays. To solve these difficulties, we combine Software Defined Networking (SDN) based central controller with fog computing in this research. SDN is a kind of network engineering that allows fog computing for intelligent and centralized network through software applications. So that SDN can be included into Fog computing for increasing network effectiveness. Hence Fog nodes send processing capabilities to the connected devices, which can dramatically reduce latencies and deliver the location services and transport protocols. Software Defined Fog Networking (SDFN) offered the adaptive central power system and basic information about the connection. Hence, we proposed a hybrid framework of SDFN- Weighed Quantum based Particle Swarm Optimization (SDFN-WQPSO) uses quantum particle motion to reach the optimal solution and achieve global convergence. Efficiency of the combinatorial optimization particle swarm, which helps to implement the fog computing system architecture proposed in the AV. Based on the simulated results, the SDFN-WQPSO approach based on SDN has the potential to significantly reduce duration and improve QoS in the SDFN system.