Performance assessment of dedicated protection schemes in multiband elastic optical networks
Year: 2026
Type: article
Abstract: Multiband technology combined with elastic optical networks (MB-EONs) has emerged as a promising resource to migrate toward sufficient capacity optical networks. Resource provisioning schemes in MB-EONs combine the advantages of a flexible grid and an ultrawide spectrum to allow more traffic admissibility. Any eventual failure may lead to huge capital and performance losses, and therefore, operators rely on resilient schemes to safeguard these enormous bandwidth-carrying resources. However, resilience comes at the cost of spectral redundancy and computational complexity. In this work, we propose and compare seven resource assignment techniques in MB-EONs using different dedicated path protection (DPP) and dedicated band protection (DBP) techniques. The seven resource assignment techniques are heuristic-based, addressing greedy (two techniques), redundancy-aware non-greedy (two techniques), and function-specific band selections (three techniques). These techniques are evaluated in eight different MB-EONs in dynamic traffic scenarios. The performance evaluation compares several parameters including bandwidth blocking, the percentage of protected and unprotected connection requests, redundancy, post-failure survivability, and carried traffic for a given requested traffic. We observe that there is a trade-off between blocking and protection performance, which impacts the achievable throughput. In particular, two DBP techniques prioritizing the S- and E-bands over the C- and L-bands and prioritizing the L- and E-bands over the C- and S-bands perform best in terms of blocking, redundancy, and time complexity. An ILP-based validation on a toy topology is also provided to confirm that the observed heuristic trends are consistent with an optimization reference.
DOI: 10.1364/jocn.592500
Institution: Universität der Bundeswehr München
Language: English
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7207825004