Joao Pedro | Searcharxiv

arXiv2026

Enhanced Scalability of Horseshoe-and-Spur Networks by Exploiting Hollow-Core Fiber

The convergence of metro and access networks into unified optical infrastructures requires cost-effective alternatives to conventional solid-core fiber (SCF). This paper examines using hollow-core fiber (HCF) with digital subcarrier multiplexing (DSCM) transceivers in horseshoe-and-spur filterless optical architectures. Leveraging HCF's ultra-low nonlinearity, we optimize amplifier placement to maximize power budgets under realistic constraints. Our results show that HCF shifts the main limitation from nonlinearity to amplifier output power, enabling up to a 20 dB spur power-budget gain over SCF. Considering balanced and unbalanced couplers, we find that increasing amplifier density boosts reach only up to a saturation point (about 11-13 units in a 5-transit-node network). A techno-economic break-even analysis of hybrid SCF-HCF deployments shows that targeted HCF use provides intermediate performance gains and can fully recover its fiber cost premium through amplifier reductions.

physics.optics↗

arXiv2025

From Data to Decision: A Multi-Stage Framework for Class Imbalance Mitigation in Optical Network Failure Analysis

Machine learning-based failure management in optical networks has gained significant attention in recent years. However, severe class imbalance, where normal instances vastly outnumber failure cases, remains a considerable challenge. While pre- and in-processing techniques have been widely studied, post-processing methods are largely unexplored. In this work, we present a direct comparison of pre-, in-, and post-processing approaches for class imbalance mitigation in failure detection and identification using an experimental dataset. For failure detection, post-processing methods-particularly Threshold Adjustment-achieve the highest F1 score improvement (up to 15.3%), while Random Under-Sampling provides the fastest inference. In failure identification, GenAI methods deliver the most substantial performance gains (up to 24.2%), whereas post-processing shows limited impact in multi-class settings. When class overlap is present and latency is critical, over-sampling methods such as the SMOTE are most effective; without latency constraints, Meta-Learning yields the best results. In low-overlap scenarios, Generative AI approaches provide the highest performance with minimal inference time.

cs.LG↗