A Priority-Based Multiobjective Design for Routing, Spectrum, and Network Coding Assignment Problem in Network-Coding-Enabled Elastic Optical Networks

Abstract

In elastic optical networks, the use of network coding (NC) represents a new dimension to further optimize spectrum efficiency, and indeed, combining NC and dedicated path protection has paved the way for achieving greater capacity efficiency, while retaining the merit of near-instantaneous recovery. In order to harness the NC benefits, a more complicated problem called routing, spectrum, and network coding assignment (RSNCA) has to be solved, and in this article, we propose a priority-based multiobjective design for the RSNCA problem aiming at maximizing the network throughput in the constrained bandwidth capacity and simultaneously minimizing the spectrum link usage for accepted demands. The multiobjective design is based on the weighting method, and we present a rigorous analysis on the impact of weight coefficients to the priority of constituent objectives. The efficacy of our design proposal is benchmarked with reference ones based on the traditional single-objective model and for both coding and noncoding approaches on various realistic topologies. It is highlighted that the application of NC brings about considerable throughput enhancement, and furthermore, the multiobjective RSNCA design is highly more efficient than the single-objective RSNCA, as up to more than 50% saving on spectrum link usage could be attained.