Enhancing Elastic Optical Network Efficiency: Advanced Resource Allocation and Optimization, and Progressive Planning
Open Access DepositedThe technology of Elastic Optical Networks (EONs) can be a reliable approach for meeting the demand of high capacity and flexibility in current backbone optical networks. Recent advancements in optical network technologies, particularly in space division multiplexing (SDM) techniques, have opened up new possibilities for enhancing service provisioning and alleviating the impending capacity crunch. The multicore fiber (MCF) technology, under the umbrella of SDM technologies, leverages fine-grained, flexible frequency grids, multiple spatial modes (fiber cores), and various advanced modulation formats to increase the capacity of optical fiber networks. However, the deployment of these technologies brings about significant challenges, primarily driven by intercore crosstalk (XT), which degrades the quality of transmission (QoT) and resource allocation. A critical problem in the MCF-based SDM EON is the Routing, Modulation, Core, and Spectrum Assignment (RMCSA) problem, which must balance several factors to ensure optimal network performance. Several RMCSA algorithms have been proposed to tackle this issue, with a notable one being our proposed Tridental Resource Assignment algorithm (TRA). TRA employs the tridental coefficient (TC) to strike a balance between spectrum utilization and XT accumulation. Given the success of TRA in efficiently assigning the resources, we present translucency aware TRA (TaTRA) which assigns resources in the translucent MCF-based SDM-EON. Nevertheless, the computational complexity of TRA is a challenge, as TC calculations are required for all possible resource choices. To address these challenges, we explore various techniques to optimize TRA. We first explore the effect of reduced resource choices on the performance of TRA. We then combine it with the exhaustive search of the optimal weights in TC with a smaller increment. Later we explore machine learning-enabled tuning of TC weights, which turns out to be a promising approach to improve the performance of TRA. By tailoring TC weights and employing multistage optimization techniques, we significantly reduce the computational overhead of TRA while maintaining or even enhancing its performance. These optimizations result in substantial reductions in bandwidth blocking probability, making TRA more practical for real-world network scenarios. In addition to designing RMCSA algorithms, we also focus on improving the balance between spectrum utilization and XT tolerance in MCF-based SDM EONs. Machine learning-aided threshold optimization strategy is proposed to enhance the performance of XT-aware RMCSA algorithms. By imposing limits on the number of lit cores on the overlapping spectrum, this strategy ensures compliance with XT constraints while maximizing spectrum efficiency, resulting in significant reductions in bandwidth blocking probability. Interestingly, the machine learning optimization works for any RMCSA algorithm as a generic optimization tool. It guarantees the improvement in the performance of any RMCSA algorithm. Furthermore, we investigate the migration of existing optical networks, such as C-band enabled Elastic Optical Networks (EONs), to newer technologies like multiband elastic optical networks (MB-EONs) and SDM-EONs. We propose an upgrade strategy called Progressive Optics Deployment and Integration for Growing Yields (PRODIGY) which ensures a smooth transition while maintaining network robustness and meeting service level agreements. PRODIGY selects the links for upgrade carefully and uses a smaller budget while increasing network lifetime. Overall, these advancements as an outcome of our contribution promise to address the challenges posed by the growing demand for network capacity effectively. These advancements are crucial for maintaining the quality of service in optical networks while accommodating the ever-increasing traffic demands.
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.