Enhancing Multipath Routing and QoS in SDNs with AI Algorithms

Software-defined networks performs a pivotal role in transforming networks from legacy to modern and automated networking. The surge in IoT service demands necessitates quality service mechanisms for user flows, tailored to specific applications. Multipath routing, a key data transmission method, uses multiple network paths to reduce traffic delay and congestion, enhancing throughput and achieving effective load balancing. This paper presents a modified artificial intelligence search algorithm (uniform cost search) to enhance multipath routing, aligning with Quality of Service (QoS) needs based on flow type. We compare our algorithm with the Deep-First Search multipath algorithm, focusing on latency, jitter, and throughput. The results highlight our algorithm's effectiveness in selecting optimal network paths for various applications, significantly improving QoS parameters.