International Journal Of Imaging

Abstract

An Integrated Approach for Monitoring and Optimizing Traffic Flow in Software-Defined Networks Using Artificial Intelligence Techniques

Author: Kranti Kumar Appari

Abstract:

This paper presents an integrated approach for monitoring traffic flow in Software-Defined Network (SDN) environments and optimizing the obtained data using artificial intelligence techniques. Traditional network monitoring approaches prove inadequate for identifying traffic bottlenecks, predicting network behavior, and optimizing routing decisions in increasingly complex SDN architectures. To address these challenges, we propose a methodology combining traffic data collection using the Floodlight controller, traffic prediction using Artificial Neural Networks (ANNs), and route optimization using novel hybrid algorithms. Our experimental results demonstrate that the proposed ANN model achieves high prediction accuracy across various network topologies, with R-squared values reaching 0.97 and Mean Absolute Percentage Error (MAPE) as low as 3.1%. Furthermore, we compare four optimization algorithms—linear search, traditional tabu search, a modified tabu search, and a novel blen