ABSTRACT
In recent years, it has become increasingly difficult to operate large networks for performing diagnostics and preventing cascading failures. Hence, there is a requirement for networks to think and learn in a nondeterministic way. This is where cognitive approach plays a vital role to overcome the shortcomings. Cognition is the ability to effectively self-regulate, learn, and evolve. Cognitive network (CN) aspects have become crucial when a system is subject to a complex and varied set of stimuli, which is certainly the case of fast-evolving Internet of Things. CNs require each node to cooperate with the data distribution
10.1 Introduction 357 10.2 Background 360
10.2.1 Cognitive Radio 360 10.2.2 Autonomic Network 361 10.2.3 Motorola FOCALE 362 10.2.4 Software-Defined Networking 363
10.3 Learning and Reasoning for CNs 363 10.4 Bio-Inspired Intelligent Network 364 10.5 Graded Cognitive Network 365 10.6 Simulation Results 370 10.7 Future Research Direction 373 10.8 Conclusion 373 References 374 Biographical Sketches 375
process and make use of information about the network scenario. To get CN working, there is a need to rethink on the architecture and protocols of the components in the global communication infrastructure. A prominent research direction looks into how to mimic nature-like mechanisms to realize smarter communication networks, which in turn can make sense of the hidden communication patterns and do the self-regulation of the topology.
