Energy Efficient RWA Strategies for WDM Optical Networks

We consider the energy minimization problem in optical networks from an algorithmic perspective. Our objective is to plan optical WDM networks so as to minimize the energy expended, by reducing the number of energy-consuming components, such as amplifiers, regenerators, add/drop terminals, optical fibers, etc. We initially present an algorithm for solving the energy-aware routing and wavelength assignment problem based on an integer linear programming formulation that incorporates energy consumption and physical impairments (through a maximum transmission reach parameter) into routing and wavelength assignment. We then present a second algorithm that decomposes the problem and uses a linear programming relaxation to address the problem in large scale networks. Simulations are performed to evaluate and compare the performance of the proposed algorithms. In previously published works, energy minimization derives mainly from the reduction of the electronic processing of the traffic and the bypass in the optical domain, while the energy consumed by the optical devices is usually neglected. We focus on the optical layer and show that energy reductions can be obtained in that layer also.