Optimization in Designing Complex Communication Networks

Abstract

AbstractComplex networks are found in real world in different areas of science, such as technological, social and biological. These networks are many times characterized by a non-trivial topology, with connection patterns among their elements that are neither purely regular nor purely random. The interesting features presented by this class of networks may be useful in improving the overall efficiency of engineered networks as computer, communication and transportation ones. There is a conjecture indicating that such complex topologies normally appear as a result of optimization processes. Optimization techniques have been applied to design complex communication networks, showing that features such as small path length, high clustering coefficient and power-law degree distribution can be achieved through optimization processes. In this chapter, models and algorithms based on optimization techniques to generate complex network topologies are discussed. We review some models, heuristics as well as exact solution approaches based on Integer Programing methods to generate topologies owning complex features.KeywordsCluster CoefficientScale Free NetworkSmall WorldSmall World NetworkAverage Path LengthThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.