Regenerator allocation for scheduled lightpath demands in translucent optical networks

Abstract

Physical layer impairments limit the maximum distance a signal can travel in the optical domain in wavelength-routed optical networks. In translucent optical networks, regenerators are made available at certain nodes, so long-haul lightpaths can undergo signal regeneration, if needed. A number of different routing and wavelength assignment (RWA) techniques for translucent networks have already been proposed in the literature, for both static and dynamic traffic. However, such techniques fail to exploit the resource sharing capabilities inherent in the scheduled traffic model (STM). Recent research has clearly established the advantages of holding-time-aware strategies for resource allocation of such scheduled (periodic) lightpath demands. In this paper, we propose a new and efficient integer linear program (ILP) based formulation for resource allocation of scheduled lightpath demands in translucent optical networks. To the best of our knowledge, this is the first holding-time-aware approach that allows optimal sharing of regenerators and other resources among non-overlapping demands to obtain a cost-effective solution. Simulation results indicate it is possible achieve significant improvements compared to existing techniques that do not consider such resource sharing.