Search for: All records

Accurate modeling of the gain spectrum in erbium-doped fiber amplifiers (EDFAs) is essential for optimizing optical network performance, particularly as networks evolve toward multi-vendor solutions. In this work, we propose a generalized few-shot transfer learning architecture based on a semi-supervised self-normalizing neural network (SS-NN) that leverages internal EDFA features—such as VOA input/output power and attenuation—to improve gain spectrum prediction. Our SS-NN model employs a two-phase training strategy comprising unsupervised pre-training with noise-augmented measurements and supervised fine-tuning with a custom-weighted MSE loss. Furthermore, we extend the framework with transfer learning (TL) techniques that enable both homogeneous (same-feature space) and heterogeneous (different-feature sets) model adaptation across booster, pre-amplifier, and ILA EDFAs. To address feature mismatches in heterogeneous TL, we incorporate a covariance matching loss to align second-order feature statistics between the source and target domains. Extensive experiments conducted across 26 EDFAs in the COSMOS and Open Ireland testbeds demonstrate that the proposed approach significantly reduces the number of measurement requirements on the system while achieving lower mean absolute errors and improved error distributions compared to benchmark methods. 

Disaggregating optical communication systems can impact physical layer control. Recent progress on multi-domain transmission control and machine-learning provide capabilities for adaptation and development of engineering rules in the field with potential benefits for disaggregated systems. 

Abstract—Optical system management software has been migrating toward software-defined networking (SDN) methods and interfaces. The increased programmability of SDN promises greater flexibility, dynamic operation, and multivendor compatibility for optical systems. However, physical layer control systems are complicated by transmission engineering requirements, including quality of transmission, optical power stability, and multidomain service guarantees. These challenges and recent commercial and research efforts to address them are examined. 

Abstract—Disaggregating optical communication systems can impact physical layer control. Recent progress on multi-domain transmission control and machine-learning provide capabilities for adaptation and development of engineering rules in the field with potential benefits for disaggregated systems.