Towards Finding an Effective Uniform and Single Point Crossover Balance for Optimization of Elastic Optical Networks

The new promising technologies in the network communication field bring the possibilities of faster and more safe information sending. However, the new technologies bring up new computation problems as well. The network flow may be organized in a better way with respect to different criteria. It is common that such problems are hard not only because of their size, but also due to their NP-complete or NP-hard nature. The Genetic Algorithms are commonly known and widely used tools, which allow to effectively solve such problems. It is usual that a Genetic Algorithm is adjusted to the solved problem. The adjustments may also take into consideration the type of crossover operator used, which is one of the most primary genetic operators. Therefore, this paper takes into consideration the NP-hard Routing and Spectrum Allocation with Joint Any cast and Unicast problem and presents the results of extensive research performed to check the influence of crossover type on the overall method effectiveness.