AI-Enabled Optimal Path Selection for QoS-Aware Routing in Software-Defined Networks

Abstract

The rapid advancement of 5G and Internet of Things (IoT) technologies, there is a growing demand for efficient network management, optimized resource allocation, and robust security mechanisms. This study investigates AI-driven approaches, emphasizing the role of SoftwareDefined Networking (SDN) and hybrid frameworks in addressing these challenges. It showcases the application of Machine Learning (ML) and Deep Reinforcement Learning (DRL) to enhance Quality of Service (QoS), streamline traffic management, and ensure network scalability. Notable innovations include the Optimized Geographic and Spectrumbased Cluster Routing (OGCSR) protocol for Vehicular Ad Hoc Networks (VANETs), the SMART framework for seamless SDN migration, and the Intelligent Anomaly-aware Network Optimization using Bidirectional Gated Recurrent Unit (IANO-BGRU) model for detecting cloud-based threats. These AI-integrated solutions demonstrate significant improvements in packet delivery rates, reduced latency, and better energy efficiency, underscoring the importance of AI-SDN convergence in developing adaptive and high-performance network infrastructures.