An auxiliary-graph-based methodology for regenerator assignment problem optimization in translucent elastic optical networks

• Published in: Optical fiber technology Vol. 53; p. 102008
• Main Authors: Cavalcante, M.A., Pereira, H.A., Chaves, D.A.R., Almeida, R.C.
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01-12-2019
• Subjects: Dynamic traffic; Elastic optical network; Physical impairment; Regenerator assignment; Regenerator placement; Regenerator placement; Physical impairment; Regenerator assignment; Elastic optical network; Dynamic traffic
• ISSN: 1068-5200; 1095-9912
• DOI: 10.1016/j.yofte.2019.102008

•We propose an auxiliary-graph-based methodology to best assess regenerator assignment.•It is based on mapping the regenerator assignment problem into an auxiliary graph.•It considers that each transparent segment in a route represents an edge.•It solves the regenerator assignment problem as if it were a routing problem.•By properly defining the edge cost function, it can replicate some heuristics.•It can also create strategies that present superior performance.
In this article, we propose an auxiliary-graph-based methodology to best assess regenerator assignment choices in translucent elastic optical networks. The proposed methodology considers that each possible transparent segment in a route is represented by an edge in an auxiliary graph. Proper edge cost functions are designed and the decision on where to use regeneration resources is performed by a standard routing algorithm in the built auxiliary graph. We propose four different edge cost functions and compare the performance, in terms of blocking probability of call requests, of some networks using our proposed methodology and other heuristics available in the literature that tackle with regenerator assignment problem in translucent elastic optical networks. The simulation scenarios involved three different network physical topologies, amplified spontaneous emission noise as physical impairment generated by boosters, in-line amplifiers and pre-amplifiers, besides losses and gains observed by the optical signals along network lightpaths. By properly defining the respective edge cost function, our proposed methodology can not only replicate some heuristics previously proposed in the literature, but also create strategies that present superior performance in all analyzed scenarios and conditions.