SDN-Enabled FiWi-IoT Smart Environment Network Traffic Classification Using Supervised ML Models

Due to the rapid growth of Internet Things (IoT), applications such as Augmented Reality (AR)/Virtual (VR), higher resolution media stream, automatic vehicle driving, smart environment and intelligent e-health applications, increasing demands for high data rates, bandwidth, low latency, quality services are every day (QoS). The management network resources IoT service provisioning is a major issue in modern communication. A possible solution this use integrated fiber-wireless (FiWi) access network. In addition, dynamic efficient configurations can be achieved through software-defined networking (SDN), an innovative programmable architecture enabling machine learning (ML) automate networks. This paper, we propose supervised traffic classification scheduling model SDN enhanced-FiWi-IoT that intelligently learn guarantee based on its QoS requirements (QoS-Mapping). We capture different non-IoT device trace files flow analyze traces extract statistical attributes (port source destination, IP address, etc.). develop robust process module framework, using these network-level classify devices. tested proposed 21 IoT/Non-IoT devices with ML algorithms results showed achieve Random Forest classifier 99% accuracy compared other techniques.