Resource Allocation in Multicore Elastic Optical Networks: A Deep Reinforcement Learning Approach

Abstract

A deep reinforcement learning (DRL) approach is applied, for the first time, to solve the routing, modulation, spectrum, and core allocation (RMSCA) problem in dynamic multicore fiber elastic optical networks (MCF-EONs). To do so, a new environment was designed and implemented to emulate the operation of MCF-EONs - taking into account the modulation format-dependent reach and intercore crosstalk (XT) - and four DRL agents were trained to solve the RMSCA problem. The blocking performance of the trained agents was compared through simulation to 3 baselines RMSCA heuristics. Results obtained for the NSFNet and COST239 network topologies under different traffic loads show that the best-performing agent achieves, on average, up to a four-times decrease in blocking probability with respect to the best-performing baseline heuristic method.

DRL-based RMSCA for SDM Networks with Core Switching in Multi-Core Fibres

“…In these works, DRL employs deep neural networks (DNNs) to extract network state information and optimise a long-term cumulative reward. In MCF-based SDM networks, the DRL-RMSCA algorithm [8] presented by Pinto-Ríos et al selects only aligned cores, failing to exploit the full core switching capability of MCF-EON. On the other hand, when developing a DRL-based RMSCA scheme able to switch between cores, handling the cardinality of the observation and the action spaces becomes challenging.…”

“…On the other hand, when developing a DRL-based RMSCA scheme able to switch between cores, handling the cardinality of the observation and the action spaces becomes challenging. Additionally, with an extremely large action space, it might not be possible to guarantee that invalid actions are not ignored by a trained DRL agent, which negatively impacts the blocking probability (BP), or will require a higher training cost [8].…”