Hybrid Regression Model for Link Dimensioning in Spectrally-Spatially Flexible Optical Networks

The latest trends in computer networks bring new challenges and complex optimization problems, one of which is link dimensioning Spectrally-Spatially Flexible Optical Networks. time-consuming calculations related to determining the objective function representing amount accepted traffic require heuristics search for good quality solutions. In this work, we address problem by proposing a hybrid regression model capable estimation. presented algorithm uses machine learning built on already evaluated solutions choosing promising ones, providing fast effective method solving considered problem. experimental evaluation conducted two representative network topologies demonstrates that proposed approach can significantly outperform other methods case EURO28 topology, while US26 it provides results comparable obtained so far.