Nokia (Germany)

The development of 5G networks has enabled diverse applications with stringent communication and computation requirements. Ensuring timely operation for these applications requires proximity to compute and network resources, such as edge servers and User Plane Functions. With control over network resources and partial ownership of compute resources, Mobile Network Operators (MNOs) can assume a larger role in service orchestration within a multi-owner edge infrastructure. However, the increasing demand for edge applications often exceeds MNO capacity. To address this, some user requests can be offloaded to third-party cloud operators, expanding capacity but incurring costs and reducing profitability. Beyond profit, energy consumption is a key concern, particularly as we transition toward 6G. Addressing high energy consumption adds complexity to resource orchestration, as it directly conflicts with the goal of maximizing profit. Moreover, in dynamic edge environments, frequent service migrations increase orchestration overhead and can negatively impact user experience. Navigating these trade-offs requires an effective optimization approach. To this end, we formulate a multi-objective optimization problem for demand placement and routing, aiming to maximize MNO profit while minimizing energy consumption and migration overhead. Given its NP-hard nature, we propose a ranked greedy heuristic to effectively balance these objectives. Simulation results show that our approach outperforms benchmark algorithms in both static and dynamic scenarios, achieving near-optimal performance and deviating less than 7% from the optimal solution.

Modern metro and access optical networks, driven by 5G and fiber-to-the-home (FTTH), demand significantly higher capacity than current state-of-the-art direct-detection solutions can provide. Consequently, high-capacity coherent transceivers deployed, for example, in backbone and datacenter interconnects links, are now expanding into these new segments. However, unlike traditional long-haul links, the primary constraint in Passive Optical Networks (PONs) is the power budget. To ensure cost-effectiveness, efficiently scale the network, and reduce the need for in-line amplifiers, these systems may need to operate at elevated power levels to overcome distribution losses. However, this approach might increase the impact of fiber Kerr nonlinearities, which are typically negligible at lower launch power levels. This study addresses the resulting challenge of accurately modeling Nonlinear Interference (NLI) for short-reach links, which traditionally has been neglected. We validate a Monte-Carlo based Enhanced Gaussian Noise (EGN) model against Split-Step Fourier Method (SSFM) and the experimental measurements in 400G shortreach transmission links. The experimental results show a promising agreement (maximum deviation error <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\approx$</tex-math></inline-formula> 0.30 dB) with the EGN model. This work confirms the EGN model as a robust and reliable tool for designing also next-generation short-reach networks, enabling more efficient network design by accounting for the interplay between NLI and transceiver noise characteristics.

The increasing complexity and capacity demands of modern optical communication networks require advanced tools for predictive maintenance and fault prevention. This study presents a systematic methodology for generating a comprehensive network failure database using Digital Twins (DTs) based on the open-source optical network simulation framework GNPy (Gaussian Noise model in Python). GNPy enables the generation of synthetic datasets to support the development of Machine Learning (ML) models for proactive fault detection and network performance optimization. By modeling a wide range of failure scenarios such as amplifier degradation, span loss variation, and component misconfigurations, GNPy enables detailed analysis of their impact on critical transmission parameters including Generalized Signal-to-Noise Ratio (GSNR), Optical Signal-to-Noise Ratio (OSNR), and channel power. The proposed methodology allows the generation of a diverse and scalable database that reflects both normal and degraded network operating conditions. This synthetic dataset provides valuable input for training ML algorithms to detect subtle performance degradations and predict potential failures before they impact running services. By leveraging this simulation-based data generation approach, network operators can significantly enhance the reliability and resilience of their transport networks. Furthermore, the integration of GNPy with data-driven ML models lays the foundation for intelligent network management systems capable of automated fault prediction and proactive maintenance strategies in future high-capacity optical networks.

Internet of Things (IoT) Mobile Network Aggregators (MNAs) are an increasingly successful paradigm for providing ubiquitous mobile Internet connectivity for Smart Devices. They leverage the 4G/5G roaming architecture offered by Mobile Network Operators (MNOs) worldwide to offer a single gateway to the Internet. In this context, promptly detecting network outages is fundamental: Mobile Network Aggregators (MNAs) do not have control over the visited network infrastructure, and they offer connectivity to a huge number of possibly unmanaged devices. However, this activity is currently a painstaking process carried out manually by network engineers, who review trends in control plane signalling messages that are forwarded in the Mobile Network Aggregators (MNAs) managed core. In this paper, we present Faro, a self-supervised solution for the automatic detection of network outages. Faro leverages a contrastive learning framework that is particularly suited for this scenario, characterised by very skewed input data, few samples, and the use of confidence scores to automatically detect changes in the input data. We evaluate Faro using a real-world dataset comprising more than 5 million control plane messages across multiple countries and over 6 months, demonstrating its ability to accurately diagnose all outage scenarios in less than 15 minutes.

We envision AI-native, self-healing optical networks that integrate LLMs, neuromorphic edge intelligence, and photonic cognition-enabling autonomous fault management and zero-touch resilience for sustainable infrastructure by 2030.

We review four years of studies on various applications of coherent point-to-multipoint transceivers enabled by digital subcarrier multiplexing. We highlight the benefits of this approach to realize truly sustainable, scalable, and flexible optical networks. ©2025 The Author(s)

We demonstrate super-rated PON downstream transmission at 100Gb/s net rate using PAM2 and PAM4 line rates ranging between 109 and 124Gb/s, exploiting an optimized LOPC code set. We showcase loss budgets >30dB for transmission equivalent to 20km at 1320nm.

Using a fast and accurate EGN model in digital subcarrier-based communication, we address the NLI noise in a horseshoe network. This method offers a computationally cheap simulation environment with the advantage of being scalable to different DWDM configurations, offering a robust solution for low-complexity networks.

Forecasting state of polarization (SOP) in aerial fibers poses several challenges due to exposure to dynamic und unpredictable environmental factors, such as wind-induced vibrations, leading to multi-timescale SOP variations. Traditional statistical and deterministic approaches may struggle to accurately model the complex, nonlinear, and multi-scale dynamics of these SOP fluctuations, making them less effective for real-world scenarios. To address these challenges, this study proposes a weather-adaptive forecasting framework combining discrete wavelet transform and artificial neural networks to predict SOP changes across varying timescales. By integrating environmental data, the model effectively captures rapid fluctuations and gradual drifts, significantly improving forecasting accuracy. Validation using field data across two locations demonstrates the model's reliability for both short-term and long-term predictions. Including environmental factors namely wind speed, temperature, and humidity improves long-term forecasting accuracy by over 64.5% compared to models taking as input SOP change data alone. Wind gusts enhance short-term forecasting under extremely windy conditions. Our proposed framework outperforms conventional techniques, offering a reliable framework for SOP change prediction and advancing proactive optical system monitoring.