Bandwidth demand adaptive link weight aware OSNR based dynamic and hybrid routing, modulation, and spectrum allocation in multi-band multiplexed elastic optical networks

Abstract

5G communication necessitates substantial improvements in optical networking to accommodate heterogeneous traffic bandwidths. The optical Routing, Modulation, and Multi-Band Spectrum Assignment (RMMBSA) algorithm offers a vital solution for resource allocation within multi-band multiplexed elastic optical networks (MB-EONs). The RMMBSA algorithm effectively manages the broadened spectrum, which includes the conventional C-band as well as additional bands such as L, S, and E, thereby enhancing spectrum utilization. This paper focuses on the spectral ranges of the C- and L-bands for resource provisioning, utilizing linear impairments, such as the optical signal-to-noise ratio (OSNR) as the criterion for band selection. The OSNR estimation is derived from the amplified spontaneous emission noise (ASE) of the path amplifiers. The substantial rise in the availability of frequency slots (FSs) for resource allocation within a multi-band optical network framework necessitates that the Software-Defined Networking (SDN) controller optimizes spectrum allocation with the help of various routing; such as static, dynamic, and hybrid approaches. The RMMBSA algorithm based on static routing has already been presented in the existing literature. This paper presents a dynamic routing-based RMMBSA algorithm, which employs a resource-aware mechanism to adjust link weights, thereby enabling the modification of routing paths in accordance with real-time traffic conditions. Furthermore, we will explore the challenges associated with the proposed technique and aim to tackle these issues by introducing a hybrid routing-based RMMBSA algorithm. The following performance metrics have been utilized to assess the effectiveness of the proposed algorithms: bandwidth blocking probability (BBP), Fractional Spectrum Utilization (FSU), link fragmentation metric (LFM), entropy-based fragmentation metric (EBFM), request blocking probability (RBP), and throughput, with BBP serving as the primary performance indicator. Simulations were conducted on the Telecom Italia (TI) and German network topologies. The results demonstrate that the proposed algorithms significantly lower the key performance metric, BBP, achieving reductions of 60.4% and 54.1% for TI and German network topologies respectively.