Complex adaptive communication networks and environments: Part 1

As modern networks grow in size, complexity and variety, the change in networks is not just terms of scale but also in the emergence of newer types of communication networks such as wireless sensor, ad-hoc, peer-to-peer (P2P), multiagent, nano-communication, mobile robot, Internet of Things (IoT), cloud-based and social networks. Various inherent nonlinearities in network operations can lead to an increase in complex emergent phenomena—phenomena whose effects are often untraceable to individual network components. Such emergent patterns can be important to understand since they can have considerable unanticipated effects on various aspects of a network. These effects can range from unanticipated traffic congestion, an unprecedented increase in communication costs, to perhaps a complete network shutdown or grid blackout. Such phenomena thus require modeling communication networks by considering them instead as artificial Complex Adaptive Systems (CAS), or generalized collectively as Complex Adaptive COmmunicatiOn Networks and environmentS (CACOONS). Recent examples of the manifestation of ideas of complexity and emergence in CACOONS include cascading failures reported in the Amazon.com cloud, effects of viral and worm infections in large networks, emergence of cascading faults in message queue–based financial transactions after New Year’s Day, network congestion and queue sizes, effects of torrent and other complex traffic on company intranets, multiplayer gaming, multiagent systems, self-organization and self-assembly in sensor and robotic communication networks and complex power networks. Because of their peculiar nature, these ideas can be better modeled, simulated, analyzed as well as visualized using techniques developed as part of modeling and simulation of living or life-like or life-inspired complex systems—specifically, techniques and ideas previously explored by numerous multidisciplinary studies in the area of CAS. Unlike other modeling and simulation paradigms, Agent-based Modeling (ABM) offers a flexible general-purpose set of techniques for modeling complex phenomena ranging from the sciences to humanities. While ABM is often used to model the dynamic behavior in CAS, Complex Network-based Modeling (CN) offers techniques to model interaction of different components using interaction datasets (especially big data). Both these paradigms have been used extensively in the modeling of social, biological, ecological, archeological and other scientific domains, and recent work has demonstrated that these paradigms can also offer a much shorter learning curve and the ability to flexibly model complex phenomena in communication networks. These techniques might be more effective for modeling and simulation of application case studies, testing of new communication protocols, investigation of problems before or after deployment or for modeling improvement in existing algorithms and hardware or for modeling communication of human or animal interaction in large-scale, hybrid, pervasive, mobile and social networks.