Multi-objective optimization of regenerator placement using Constraint Programming

In translucent optical networks, connections bypass most intermediate nodes transparently. However, their signal quality is affected by the traveled distance and/or the bypass operations. Hence, they might require optical-electrical-optical conversions at some intermediate nodes, in order to regenerate the carried signal. In this paper, we evaluate several approaches to deal with connection routing and regenerator placement within a translucent network. We present our Constraint Programming based multi-objective optimization model which accepts three criteria: total routing cost, total number of regeneration operations, and number of nodes performing regeneration. Our results show that this approach provides solutions in a reasonable running time for problem instances of practical interest. We discuss our method in the context of the state-of-the-art and show that optimization based on one criterion only misses the possibility to improve the solution in respect to the other aspects.