An Intelligent Traffic Classification Based Optimized Routing in SDN-IoT: A Machine Learning Approach

Abstract

Due to speedy increase in IoT devices and its QoS requirements, providing networks solutions to meet this demand has become a major research issue. Providing fast and reliable routing paths based on the QoS requirement of IoT device is very vital. Software defined networking is one of the most current interesting development in the field of research. A new paradigm, SDN-IoT, leveraging the advantages of SDN architecture on IoT networks have been proposed to improve network quality. Also, application of artificial intelligence (AI) in SDN for traffic engineering is widely researched. In this work, we first propose a machine learning based traffic load classification into the traffic’s QoS requirements. Then, a deep learning route optimization model based on the traffic classification is proposed. The model chooses the route that meets the QoS demands like latency of the identified traffic. The simulation results show that our proposed solutions perform very well and better than some significant works in the same area.