Searcharxiv⌕ Search

arXiv subjects

Massimo Tornatore

Publications and source records attributed to Massimo Tornatore.

At least 19 recordsLinked to original sources
arXiv2026

Measurement-Device Placement in MDI-QKD Networks with Minimal Trusted Relays

Measurement-Device-Independent Quantum Key Distribution (MDI-QKD) removes detector side-channel vulnerabilities by delegating measurements to an untrusted relay that, when shared by several user pairs, acts as a Bell-state measurement (BSM) hub. Channel loss still limits the reach of MDI-QKD links, so long-range services rely on trusted relays for key forwarding. As MDI-QKD moves toward metropolitan deployment, a key network-planning question arises: how should BSM hubs be placed on existing fiber infrastructure to minimize the use of trusted relays? In this work, we formalize this challenge as the MDI-QKD Hub Deployment (MHD) problem. We first prove that the MHD problem is NP-hard, and then formulate it as a Mixed-Integer Linear Programming (MILP). To the best of our knowledge, this is the first formulation for MDI-QKD network planning that captures the structural features specific to MDI-QKD, namely a shared BSM hub, two-link user-to-hub routing, and loss-balancing constraints, while jointly determining hub placement, user-to-hub assignment, and trusted-relay demand allocation. Our solution accounts for realistic budget and capacity constraints, including practical insights from real-world deployments of commercial MDI-QKD solutions. Numerical evaluations on three metropolitan topologies (12 to 50 nodes) at realistic geographic distances show that topology structure governs trusted-relay demand: hub placement alone eliminates all trusted relays in compact urban meshes, while in sparse topologies, relaxing the loss-balancing constraint removes the dominant source of trusted-relay usage at no additional infrastructure cost. The key-rate threshold at which trusted relays first appear shifts monotonically with the network diameter, i.e., the largest shortest-path distance between any node pair in fiber kilometers, delineating the feasibility boundary of pure MDI-QKD deployment.

cs.NI↗
arXiv2026

LLMs and Optical Networks: A Symbiotic Relationship

This paper explores the emerging symbiosis between LLMs and optical networks. Massive LLMs require geo-distributed training, which demands advanced optical transport capabilities that require new key technical enablers, as WAN-aware CCL algorithms, ZR+ pluggables, and Hollow Core Fibers. Conversely, LLMs also enable new forms of autonomous network management.

cs.NI↗
arXiv2026

Phi-Actor-Critic: Steering General-Sum Games to Pareto-Efficient Correlated Equilibria

Real-world multi-agent systems, from traffic coordination to resource allocation, are often modeled as general-sum games where individual incentives conflict with collective welfare. In these settings, the central challenge is not merely finding an equilibrium, but selecting socially desirable outcomes among many suboptimal Nash equilibria. Standard deep multi-agent reinforcement learning (MARL) methods struggle with this problem, as value-decomposition approaches are constrained by monotonicity assumptions and policy-gradient methods often converge to stable but socially inefficient equilibria. To address this limitation, we propose $\Phi$-Actor-Critic ($\Phi$-AC), a framework that leverages swap regret minimization to steer learning toward high-welfare correlated equilibria (CE). To make counterfactual regret estimation tractable in deep MARL, $\Phi$-AC employs a centralized attention critic that predicts vector-valued regrets in a single forward pass, avoiding computationally expensive counterfactual simulations. We further introduce a Lagrangian-based equilibrium selection mechanism that optimizes social welfare while enforcing stability through regret constraints. Experiments on matrix games, Multi-Agent Particle Environments (MPE), and the Melting Pot Harvest scenario demonstrate that $\Phi$-AC learns efficient and stable coordination strategies across diverse mixed-motive settings while maintaining high collective return and competitive fairness.

cs.MA↗
arXiv2025

Holoscope: Open and Lightweight Telescope & Honeypot Platform

The complexity and scale of Internet attacks call for distributed, cooperative observatories capable of monitoring malicious traffic across diverse networks. Holoscope is an open, lightweight, and cloud-native platform designed to simplify the deployment and management of telescope (passive) and honeypot (active) sensors. Built upon K3s and WireGuard, Holoscope offers secure connectivity, automated sensor onboarding, and resilient operation even in resource-constrained environments. Through modular design and Infrastructure-as-Code principles, it supports dynamic sensor orchestration, automated recovery, and data processing. We build, deploy, and operate Holoscope across multiple institutions and cloud networks in Europe and Brazil, enabling unified visibility into large-scale attack phenomena while maintaining ease of integration and security compliance.

cs.DC↗
arXiv2025

Routing and Wavelength Assignment with Minimal Attack Radius for QKD Networks

Quantum Key Distribution (QKD) can distribute keys with guaranteed security but remains susceptible to key exchange interruption due to physical-layer threats, such as high-power jamming attacks. To address this challenge, we first introduce a novel metric, namely Maximum Number of Affected Requests (maxNAR), to quantify the worst-case impact of a single physical-layer attack, and then we investigate a new problem of Routing and Wavelength Assignment with Minimal Attack Radius (RWA-MAR). We formulate the problem using an Integer Linear Programming (ILP) model and propose a scalable heuristic to efficiently minimize maxNAR. Our approach incorporates key caching through Quantum Key Pools (QKPs) to enhance resilience and optimize resource utilization. Moreover, we model the impact of different QKD network architectures, employing Optical Bypass (OB) for optical switching of quantum channels and Trusted Relay (TR) for secure key forwarding. Moreover, a tunable parameter is designed in the heuristic to guide the preference for OB or TR, offering enhanced adaptability and dynamic control in diverse network scenarios. Simulation results confirm that our method significantly outperforms the baseline in terms of security and scalability.

quant-ph↗
arXiv2025

Reliability and Availability in Virtualized Networks: A Survey on Standards, Modeling Approaches, and Research Challenges

The rise of Network Function Virtualization (NFV) has transformed network infrastructures by replacing fixed hardware with software-based Virtualized Network Functions (VNFs), enabling greater agility, scalability, and cost efficiency. Virtualization increases the distribution of system components and introduces stronger interdependencies. As a result, failures become harder to predict, monitor, and manage compared to traditional monolithic networks. Reliability, i.e. the ability of a system to perform regularly under specified conditions, and availability, i.e. the probability of a system of being ready to use, are critical requirements that must be guaranteed to maintain seamless network operations. Accurate modeling of these aspects is crucial for designing robust, fault-tolerant virtualized systems that can withstand service disruptions. This survey focuses on reliability and availability attributes of virtualized networks from a modeling perspective. After introducing the NFV architecture and basic definitions, we discuss the standardization efforts of the European Telecommunications Standards Institute (ETSI), which provides guidelines and recommendations through a series of standard documents focusing on reliability and availability. Next, we explore several formalisms proposed in the literature for characterizing reliability and availability, with a focus on their application to modeling the failure and repair behavior of virtualized networks through practical examples. Then, we overview numerous references demonstrating how different authors adopt specific methods to characterize reliability and/or availability of virtualized systems. Moreover, we present a selection of the most valuable software tools that support modeling of reliable virtualized networks. Finally, we discuss a set of open problems with the aim to encourage readers to explore further advances in this field.

cs.NI↗
arXiv2025

Vertical Federated Learning for Failure-Cause Identification in Disaggregated Microwave Networks

Machine Learning (ML) has proven to be a promising solution to provide novel scalable and efficient fault management solutions in modern 5G-and-beyond communication networks. In the context of microwave networks, ML-based solutions have received significant attention. However, current solutions can only be applied to monolithic scenarios in which a single entity (e.g., an operator) manages the entire network. As current network architectures move towards disaggregated communication platforms in which multiple operators and vendors collaborate to achieve cost-efficient and reliable network management, new ML-based approaches for fault management must tackle the challenges of sharing business-critical information due to potential conflicts of interest. In this study, we explore the application of Federated Learning in disaggregated microwave networks for failure-cause identification using a real microwave hardware failure dataset. In particular, we investigate the application of two Vertical Federated Learning (VFL), namely using Split Neural Networks (SplitNNs) and Federated Learning based on Gradient Boosting Decision Trees (FedTree), on different multi-vendor deployment scenarios, and we compare them to a centralized scenario where data is managed by a single entity. Our experimental results show that VFL-based scenarios can achieve F1-Scores consistently within at most a 1% gap with respect to a centralized scenario, regardless of the deployment strategies or model types, while also ensuring minimal leakage of sensitive-data.

cs.NI↗
arXiv2025

Multi-Failure Localization in High-Degree ROADM-based Optical Networks using Rules-Informed Neural Networks

To accommodate ever-growing traffic, network operators are actively deploying high-degree reconfigurable optical add/drop multiplexers (ROADMs) to build large-capacity optical networks. High-degree ROADM-based optical networks have multiple parallel fibers between ROADM nodes, requiring the adoption of ROADM nodes with a large number of inter-/intra-node components. However, this large number of inter-/intra-node optical components in high-degree ROADM networks increases the likelihood of multiple failures simultaneously, and calls for novel methods for accurate localization of multiple failed components. To the best of our knowledge, this is the first study investigating the problem of multi-failure localization for high-degree ROADM-based optical networks. To solve this problem, we first provide a description of the failures affecting both inter-/intra-node components, and we consider different deployments of optical power monitors (OPMs) to obtain information (i.e., optical power) to be used for automated multi-failure localization. Then, as our main and original contribution, we propose a novel method based on a rules-informed neural network (RINN) for multi-failure localization, which incorporates the benefits of both rules-based reasoning and artificial neural networks (ANN). Through extensive simulations and experimental demonstrations, we show that our proposed RINN algorithm can achieve up to around 20 higher localization accuracy compared to baseline algorithms, incurring only around 4.14 ms of average inference time.

cs.NI↗
arXiv2024

Increasing Information-Carrying Capacity by Exploiting Diverse Traffic Characteristics in Multi-Band Optical Networks

Efficient network management in optical backbone networks is crucial for handling continuous traffic growth. In this work, we address the challenges of managing dynamic traffic in C- and C+L-band optical backbone networks while exploring application flexibility, namely the compressibility and delayability metrics. We propose a strategy, named Delay-Aware and Compression-Aware (DACA) provisioning algorithm, which reduces blocking probability, thereby increasing information-carrying capacity of the network compared to baseline strategies.

cs.NI↗
arXiv2024

Cost and Power-Consumption Analysis for Power Profile Monitoring with Multiple Monitors per Link in Optical Networks

As deploying large amounts of monitoring equipment results in elevated cost and power consumption, novel low-cost monitoring methods are being continuously investigated. A new technique called Power Profile Monitoring (PPM) has recently gained traction thanks to its ability to monitor an entire lightpath using a single post-processing unit at the lightpath receiver. PPM does not require to deploy an individual monitor for each span, as in the traditional monitoring technique using Optical Time-Domain Reflectometer (OTDR). In this work, we aim to quantify the cost and power consumption of PPM (using OTDR as a baseline reference), as this analysis can provide guidelines for the implementation and deployment of PPM. First, we discuss how PPM and OTDR monitors are deployed, and we formally state a new Optimized Monitoring Placement (OMP) problem for PPM. Solving the OMP problem allows to identify the minimum number of PPM monitors that guarantees that all links in the networks are monitored by at least $n$ PPM monitors (note that using $n>1$ allows for increased monitoring accuracy). We prove the NP-hardness of the OMP problem and formulate it using an Integer Linear Programming (ILP) model. Finally, we also devise a heuristic algorithm for the OMP problem to scale to larger topologies. Our numerical results, obtained on realistic topologies, suggest that the cost (and power) of one PPM module should be lower than 2.6 times that of one OTDR for nation-wide and 10.2 times for continental-wide topology.

cs.NI↗
arXiv2024

Poster: Flexible Scheduling of Network and Computing Resources for Distributed AI Tasks

Many emerging Artificial Intelligence (AI) applications require on-demand provisioning of large-scale computing, which can only be enabled by leveraging distributed computing services interconnected through networking. To address such increasing demand for networking to serve AI tasks, we investigate new scheduling strategies to improve communication efficiency and test them on a programmable testbed. We also show relevant challenges and research directions.

cs.NI↗
arXiv2024

Throughput Maximization in Multi-Band Optical Networks with Column Generation

Multi-band transmission is a promising technical direction for spectrum and capacity expansion of existing optical networks. Due to the increase in the number of usable wavelengths in multi-band optical networks, the complexity of resource allocation problems becomes a major concern. Moreover, the transmission performance, spectrum width, and cost constraint across optical bands may be heterogeneous. Assuming a worst-case transmission margin in U, L, and C-bands, this paper investigates the problem of throughput maximization in multi-band optical networks, including the optimization of route, wavelength, and band assignment. We propose a low-complexity decomposition approach based on Column Generation (CG) to address the scalability issue faced by traditional methodologies. We numerically compare the results obtained by our CG-based approach to an integer linear programming model, confirming the near-optimal network throughput. Our results also demonstrate the scalability of the CG-based approach when the number of wavelengths increases, with the computation time in the magnitude order of 10 s for cases varying from 75 to 1200 wavelength channels per link in a 14-node network. Code of this publication is available at github.com/cchen000/CG-Multi-Band.

cs.NI↗
arXiv2021

Maximizing Revenue with Adaptive Modulation and Multiple FECs in Flexible Optical Networks

Flexible optical networks (FONs) are being adopted to accommodate the increasingly heterogeneous traffic in today's Internet. However, in presence of high traffic load, not all offered traffic can be satisfied at all time. As carried traffic load brings revenues to operators, traffic blocking due to limited spectrum resource leads to revenue losses. In this study, given a set of traffic requests to be provisioned, we consider the problem of maximizing operator's revenue, subject to limited spectrum resource and physical layer impairments (PLIs), namely amplified spontaneous emission noise (ASE), self-channel interference (SCI), cross-channel interference (XCI), and node crosstalk. In FONs, adaptive modulation, multiple FEC, and the tuning of power spectrum density (PSD) can be effectively employed to mitigate the impact of PLIs. Hence, in our study, we propose a universal bandwidth-related impairment evaluation model based on channel bandwidth, which allows a performance analysis for different PSD, FEC and modulations. Leveraging this PLI model and a piecewise linear fitting function, we succeed to formulate the revenue maximization problem as a mixed integer linear program. Then, to solve the problem on larger network instances, a fast two-phase heuristic algorithm is also proposed, which is shown to be near-optimal for revenue maximization. Through simulations, we demonstrate that using adaptive modulation enables to significantly increase revenues in the scenario of high signal-to-noise ratio (SNR), where the revenue can even be doubled for high traffic load, while using multiple FECs is more profitable for scenarios with low SNR.

cs.NI↗