Hierarchical Traffic Optimization in SDN with Graph NeuralNetworks and PSO Techniques

Software-Defined Networking (SDN) has revolutionized how networks are managed andoptimized. However, as SDNs scale in size and complexity, traffic optimization becomesa challenging task. This paper presents a novel approach to hierarchical traffic optimizationin SDNs using Graph Neural Networks (GNNs) and Particle Swarm Optimization (PSO).By combining the topological learning capabilities of GNNs with the global optimizationpower of PSO, the proposed method aims to improve network performance in terms ofthroughput, latency, load balancing, and quality of service (QoS). The model is evaluatedthrough extensive simulations, showing significant improvements over conventional SDNtraffic management methods

کلیدواژه ها:

Software-Defined Networking (SDN) ، Traffic Optimization ، Graph NeuralNetworks (GNNs) ، Particle Swarm Optimization (PSO) ، Hierarchical Optimization ، LoadBalancing ، QoS ، Routing