Guest Editorial Introduction to the Special Section on Cognitive Software Defined Networks and Applications

The papers in this special section focus on cognitive software defined networks and applications. Next generation networks (NGNs) are xpected to utilize internal and external sources of data through information and wireless communication techniques. Particularly, the demand for autonomic network management, orchestrations and optimization is as intense as ever, even though significant research has been needed. Software Defined Networks (SDNs) have been proposed to address QoS requirements for NGNs including high throughput, high mobility, low latency, heterogeneity and scalability. SDN has improved the user experience by providing high-performance communications between the network nodes, reconstructing the network structure, and optimizing the networking coverage, system security, communication latency, etc. The control intelligence is moved out of devices in a logically centralized controller, which interacts with data plane devices through standard interfaces. However, the existing applications in the SDN attract more attention to develop new learning algorithms, enhanced protocols and is even used in sensor power line communication for data transmission. The Cognitive Learning algorithms are the best solution to some particular applications. The cognitive software defined network (CSDN) presents to combine the efficiencies of SDN with new cognitive learning algorithms and enhanced protocols to automatize SDN. Its research and implementation are based on autonomic network management and control concepts. Such a combination of SDNs with autonomic frameworks and cognitive algorithms is better to solve the issues of traditional SDNs. This architecture of CSDN enables up-to-date control schemes to be developed and deployed to enable new smart networking services.