Progressive network recovery in optical core networks

Abstract

Disasters hit telecommunication structures every year. As Earth becomes more disaster-prone, telecom networks, and especially optical networks due to their high capacity, should be designed to prepare against disasters. Although disaster resiliency for optical core networks has been subject of significant recent research, damages on the network infrastructure due to the direct or indirect impacts of disasters are usually inevitable. In fact, even if undamaged parts of the network may still be functional, soon the survived resources might get exhausted to serve the residual traffic from unaffected regions. Besides, traffic from affected regions should be recovered quickly to help survivors and rescue teams to recover at least the essential communication services. Full recovery may take weeks and multi-stage progressive-recovery plans are needed. Thus, efficient recovery methods for optical core networks should also be designed. There are only a few research on progressive network recovery problem, mostly dealing with generic IP transport networks. In this work, we investigate the problem of progressive network recovery with a specific focus on the role of different optical network architectures, namely opaque, transparent, and elastic optical networks. We formulate the problem into a set of Mixed Integer Linear Programming (MILP) models. The numerical examples show that optical networks that exploit recent technological advances, i. e., elastic optical networks, perform better than classical architectures (opaque and transparent optical networks) in terms of traffic recovered, due to its higher spectrum-utilization efficiency. The numerical examples also showed that pre-deploying some spare transponders can significantly increase network recovery performance.