Cognitive algorithm to solve the impairment-aware virtual topology design problem in reconfigurable optical networks

A key problem in current optical telecommunication networks is that of virtual topology design, where a set of optical connections (or lightpaths) are established between network nodes in order to accommodate traffic demands. Those virtual topologies are designed with the aim of optimizing a set of network parameters while complying with quality of transmission requirements. In this paper, we show the advantages of using cognition when solving that problem. In particular, we present two multiobjective genetic algorithms (one of them enhanced with an additional cognitive technique) and evaluate their performance under realistic traffic demands. Results show that the new method enhanced with cognition obtains more and better results than the other method.