Defragmentation-Aware Multilayer Route and Spectrum Assignment in Elastic Optical Network

Integrating Optical Transport Networks (OTNs) into multilayer Elastic Optical Networks (EONs) enhances data transmission efficiency but introduces significant challenges in routing and spectrum allocation, particularly when Ethernet connections are encapsulated within OTN payloads. To address these challenges, we develop an Integer Linear Programming (ILP) model tailored for dynamic traffic scenarios to optimize Route and Spectrum Assignment (RSA) in multilayer EONs under OTN constraints. Recognizing the scalability limitations of the ILP model, we propose a novel heuristic algorithm based on a Collapsed Auxiliary Graph (CAG) that abstracts the multilayer network into a single layer. We employ a label-setting approach to solve the constrained shortest path problem on the CAG. This method effectively prevents spectrum interference and controlling latency along demand paths. It also enables fine-tuning of CAG edge weights to implement various traffic engineering policies. Simulation results demonstrate that our heuristic effectively reduces the blocking ratio and improves spectrum utilization while balancing network efficiency and minimizing OTN switching usage. Our findings provide valuable insights into the impact of individual layer configurations on overall network performance, highlighting the effectiveness of our approach in optimizing multilayer EONs with OTN integration.