Effective Worst-Case Crosstalk Estimation for Dynamic Translucent SDM Elastic Optical Networks

We focus on dynamic crosstalk-aware lightpath provisioning in translucent space-division multiplexing (SDM) elastic optical networks, in which spectral super-channels are carried over multi-core fibers (MCFs), distance-adaptive transmission is considered, and back-to-back (B2B) signal regeneration is applied. We aim at establishing lightpaths for connection requests in such a way that the inter-core crosstalk (XT) impairment affecting the quality of lightpaths in the network does not exceed allowable threshold levels. To this end, we propose and analyze a method that is based on per-core static/worst-case XT estimation. This method is implemented in a dynamic XT-aware routing, spatial mode, and spectrum allocation (RSSA) algorithm, which is used to solve the lightpath provisioning problem in an efficient way, such that the bandwidth blocking probability is minimized. The performed extensive simulation experiments in realistic network scenarios show a significant impact of this method on network performance and motivate for its application instead of commonly used per-fiber worst-case XT estimation.