Software Defined Intelligent Networks for Fairness and Quality of Experience Prediction and Evaluation

This thesis addresses the challenges posed by the increasing demand for online distribution of high-quality and high-throughput content. The growing number of media applications competing for network resources results in a significant impact on network efficiency and the QoE. In a multi-user multi-device environment, measuring and maintaining perceivable user feedback and fairness becomes as critical as achieving the QoE on individual user applications. To overcome these challenges, this thesis proposes novel framework designs using programmable networks such as SDN. The framework automates the measuring and maintenance of perceivable user feedback and fairness, leading to improved QoE and network efficiency. The proposed framework involves the use of ML techniques to build a prediction model based on subjective user experiments. This eliminates the need for physical experiments and automates the process of predicting QoE. Adaptive streaming over a software-defined network environment is also being examined in the framework, evaluating and studying the media streams, aspects affecting the stream, and the network. By analyzing the network's features and their direct relationship with the perceived QoE, the framework improves the network's optimization and reduces the discrepancy of QoE across user devices. Finally, this thesis discusses application- and human-level fairness over networked multimedia applications and how such fairness can be managed through novel network designs using programmable networks such as SDN. By achieving both fairness and QoE, the proposed framework can provide a comprehensive solution to the challenges posed by the increasing demand for online distribution of high-quality and high-throughput content.