Abstract:
Throughput optimization and traffic engineering in Wavelength-Division Multiplexing (WDM) networks are usually treated as mono-objective optimization problems. In this pa...Show MoreMetadata
Abstract:
Throughput optimization and traffic engineering in Wavelength-Division Multiplexing (WDM) networks are usually treated as mono-objective optimization problems. In this paper, we provide a multi-objective Integer Linear Program (ILP) for the joint throughput optimization and traffic engineering problem. By simultaneously i) maximizing the throughput, ii) minimizing the resource consumption, and iii) balancing the traffic load, we demonstrate that better solutions than those of mono-objective approaches are obtained. We also present a distributed heuristic algorithm, which upper-bounds the per-route resource consumption and maximizes the throughput. Simulation results validate the proposed model and heuristic algorithm. Additionally, we present an ILP formulation for another well-known problem such as routing and wavelength assignment (RWA), and discuss the impact of modeling it as a multi-objective problem.
Published in: 2010 IEEE International Conference on Communications
Date of Conference: 23-27 May 2010
Date Added to IEEE Xplore: 01 July 2010
ISBN Information: