A NEW Heuristically Optimized Power-Law Network TOPOLOGY DESIGN OF COMLEX NETWORKS AT ROUTER-LEVEL

Significant efforts have been made to understand the topological structures of underlying complex networks. The key is studying of the graph-theoretic properties of the corresponding connectivity structures and describing the different systems in a statistical sense. The structure of complex networks has been widely described as scale-free networks generated by the preferential attachment model. HoWever, these models do not take into account the more detailed description of the underlying topological physical structure observed in real networks. In this paper, we propose a new synthetic model for designing of Internet s router- level topology based on Heuristically Optimized Tradeoffs (HOT) concept. The tradeoff is between system throughput and likelihood-related topology metric (s-metric) subject to economic link costs and router technology constraints that are crucial when designing the synthetic system model. We propose a new edge rewiring/addition process for a small-world model with tunable parameters to address the aforementioned issues for high variability in backbone core structure connectivity. Our proposal approach can reproduce the low- likelihood topology metric (topological disassortativity of real networks) and can satisfy the small-world effect at the same time to achieve reasonably good network functional requirements. We show that our proposal approach yields synthetic network model that not only introduce power-law networks at router-level topologies, but is also consistent with engineering intuition and networking reality that their throughput aspects and robustness properties are concerned.