A Distributed SDN Controller for Distributed IoT

Over the past decade, Internet of Things (IoT) has undergone a paradigm shift away from centralized cloud computing towards edge computing. Hundreds billions things are estimated to be deployed in rapidly advancing IoT paradigm, resulting enormous amounts data. Transmitting all these data recently proven performance bottleneck, as it causes many network issues relating latency, power consumption, security, privacy, etc. However, existing paradigms do not use devices for decision making. The distributed intelligence could strengthen several ways by distributing decision-making tasks within network, rather than sending central server. In this approach, computational and shared among devices. To achieve efficient distribution utilization resources, is necessary integrate transport control with resources provide dynamic services. This challenge can overcome equipping an gateway required extract information raw decide whether actuate itself or outsource complex cloud. Distributed refers collaboration between edge. context, we first introduce SDN-based architecture that enables gateways perform processing dynamically at based on current state resources. Next, propose algorithm selecting clients training process Federated Learning Applications, context network. last step, deploy Applications our architecture, using conduct comprehensive detailed evaluation system perspectives.