A Fuzzy Logic Approach To Increase Quality of Service in Software Defined Networking

It has been a crucial task to provide and maintain an acceptable level of Quality of Service (QoS) for delivering multimedia services over Internet. Different media types are being impacted by resource limitation and various management policies in different ways. For instance, while audio services ask for a path with minimum communication latency, video streams should be routed over network links with high capacity. However, in today’s networks, communication flows that are directed towards the same destination follow the same network link which may result in inefficient resource utilisation and effectively degrading the overall QoS in the network. This paper proposes a fuzzy logic approach in which path’s QoS measurements (e.g. packet loss rate or link delay) are received as input parameters and path weights are determined for all communication paths between a source and destination. Such an approach can be beneficial for Software Defined Networking (SDN) in which a control plane orchestrates setting up the communication paths between end-users. In our proposed Fuzzy Logic Enabled-SDN (FLE-SDN) approach, the SDN controller periodically collects the QoS measurements and dynamically routes service flows (e.g. audio, video and data) to the most suitable paths in order to fulfill the QoS requirements of specific service flows. We have used MATLAB to analyse the performance of the proposed (FLE-SDN) approach. Our preliminary obtained results show that our proposed fuzzy logic control system can assist SDN controllers to better determine a communication path for a specific service flow and thus, offering better QoS.