cs.LGNov 11, 2026

Continual Event Recognition for Autonomous Optical Network Monitoring

Authors: Khouloud Abdelli

Organizations: Advanced Photonics Department, Nokia Bell Labs, Stuttgart, Germany

Abstract

Autonomous optical networks require monitoring systems capable of adapting to evolving disturbances without repeated retraining or service interruption. We propose a continual event recognition framework for state-of-polarization (SOP)-based optical monitoring based on prototype representations, open-set recognition, and few-shot prototype registration. The framework supports recognition of known disturbances, detection and integration of novel events, and investigates whether overlapping disturbances can be inferred through compositional reasoning from existing prototypes. Experimental validation on a 2600-km ROADM-based optical testbed demonstrates 99.82% closed-set recognition accuracy, successful adaptation to novel disturbances without encoder retraining, and reveals that overlapping optical disturbances exhibit varying degrees of latent compositionality. While some overlaps can be recognized in a zero-shot manner through prototype composition, others require explicit few-shot registration. These results establish continual event recognition as a promising paradigm for future autonomous optical network monitoring systems.

Explore similar work

Jun 9, 2026

A Unified Siamese Learning Framework for Zero-Day Anomaly Detection and Classification in Optical Networks

A multi-similarity Siamese neural network unifies zero-day anomaly detection and one-shot classification in optical networks, achieving over 99% accuracy and instant adaptability across lightpaths and unseen anomaly types without any retraining.
Jun 29, 2026

Hybrid Active-Online Learning Framework for Label-Efficient Concept Drift Adaptation in Optical Network Failure Detection

We propose a hybrid active-online learning framework for label-efficient concept drift adaptation in optical network failure detection. Using margin-based selective labeling, our method achieves nearceiling accuracy and AUC scores while querying only 3.4% of streaming samples, with negligible latency overhead compared to static inference.
Apr 20, 2026

Variational Autoencoder Domain Adaptation for Cross-System Generalization in ML-Based SOP Monitoring

Machine learning (ML) models trained to detect physical-layer threats on one optical fiber system often fail catastrophically when applied to a different system, due to variations in operating wavelength, fiber properties, and network architecture. To overcome this, we propose a Domain Adaptation (DA) framework based on a Variational Autoencoder (VAE) that learns a shared representation capturing event signatures common to both systems while suppressing system-specific differences. The shared encoder is first trained on the combined data from two distinct optical systems: a 21 km O-band dark-fiber testbed (System 1) and a 63.4 km C-band live metro ring (System 2). The encoder is then frozen, and a classifier is trained using labels from an individual system. The proposed approach achieves 95.3% and 73.5% cross-system accuracy when moving from System 1 to System 2 and vice versa, respectively. This corresponds to gains of 83.4% and 51% over a fully supervised Deep Neural Network (DNN) baseline trained on a single system, while preserving intra-system performance.