An adaptive model for spectrum assignment in elastic optical networks

Abstract

For an elastic optical network (EON), considering cases of various lightpaths pairs requiring different actual guard bands, we propose an adaptive spectrum assignment (ASA) model which formulates the problem such that, it is possible to investigate as to how spectrum resources can be efficiently assigned in the considered cases. Initially, ASA is formulated as an integer linear program (ILP), followed by its simplification to an ordered-ASA (O-ASA) problem which is solvable in polynomial time. To solve ASA in tractable time, we propose a dual stage heuristic algorithm in which, an initial solution is obtained in the first stage, followed by a second stage which resorts to the use of random optimisation for improving the initial solution's quality. We conduct simulations considering various graph types and realistic network topology(s), and the obtained results show that, in comparison to ILP model, the proposed dual stage heuristic algorithm finds almost optimal solutions for the ASA.