Energy-Efficient Fast Configuration of Flexible Transponders and Grooming Switches in OFDM-Based Elastic Optical Networks

Abstract

We investigate the problem of energy-efficient resource allocation constrained to quality of service and physical requirements in orthogonal frequency division multiplexing (OFDM)-based elastic optical networks and propose a fast two-stage algorithm to solve it. The first stage of the proposed algorithm deals with routing, traffic grooming, and traffic ordering and mainly aims at minimization of the number of deployed optical amplifiers and transponders. We provide an integer linear program for routing and traffic grooming and propose a heuristic procedure, which yields its near-optimum solution in a shorter runtime. In the second stage, we optimize transponder parameters to minimize total transponder power consumption. We show how signal-to-noise ratio and transponder power consumption are represented by convex expressions and use the results to provide a convex formulation for optimizing transponder parameters. Unlike the conventional formulations, we consider transmitted optical power as an optimization variable tuned for each lightpath and show how this improves power consumption of different network elements. Simulation results demonstrate that the proposed algorithm for routing and traffic grooming is around 2 orders of magnitude faster than its equivalent integer linear formulation and reduces network power consumption more than 9% compared with the scenario in which no traffic grooming is applied. It is also shown that our convex formulation for transponder parameter assignment is 1 order of magnitude faster than its mixed-integer nonlinear counterpart and reduces total transponder power consumption more than 13% compared with a fixed transponder configuration scheme. Furthermore, we investigate the effect of adaptive modulation assignment on power consumption and show it is not necessary to have complex transponders supporting a high number of modulation formats. We also analyze the impact of transponder capacity on traffic grooming and investigate the inherent trade-off between capital expenditures and operational expenditures in terms of the capacity and design complexity of the transponders.