Towards Intent Assurance: A Traffic Prediction Model for Software-Defined Networks

In modern networks, the increasing complexity and dynamic traffic patterns pose significant challenges to maintaining optimal performance and detecting traffic anomalies that can degrade network performance and lead to violations of Service Level Agreements (SLAs). This paper presents a traffic prediction model designed to prevent congestion in Software-Defined Networks (SDN) as part of the assurance operations within an Intent-Based Networking (IBN) framework. The proposed model leverages real-time network data and machine learning techniques to forecast traffic patterns, enabling proactive congestion management and resource optimization. By predicting potential bottlenecks, the model facilitates dynamic adjustments through load balancing to ensure consistent Quality of Service (QoS) and adherence to network intents. Integrated into the IBN framework, the model enhances both network reliability and user satisfaction.