Interference detection in spectrum-blind multi-user optical spectrum as a service

2025-04-02

Journal of Optical Communications and Networking

AI Summary

In elastic optical networks supporting Optical Spectrum-as-a-Service (OSaaS) for multi-tenant spectrum sharing, accurately localizing interference sources—arising from fiber nonlinearity and amplifier crosstalk—remains challenging due to the absence of fine-grained spectral visibility. This paper proposes the first spectrum-blind interference attribution framework: it performs real-time interference detection and user-level attribution without requiring full-spectrum sensing or tenant-specific spectral details, leveraging only coarse-grained channel power measurements and optical-layer performance monitoring data. The method integrates a lightweight supervised learning model, domain-informed feature engineering, and fusion of heterogeneous monitoring signals. Evaluated on the Open Ireland production testbed (190 km fiber link with three co-located tenants), it achieves 90.3% accuracy in interference source classification. This approach overcomes the conventional reliance on high-resolution spectral analysis, enabling secure multi-tenancy and operational isolation in open line systems while significantly reducing service interruption duration and fault-resolution latency.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Search and Optimization: Distributed Search
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

With the growing demand for high-bandwidth, low-latency applications, optical spectrum as a service (OSaaS) is of interest for flexible bandwidth allocation within elastic optical networks (EONs) and open line systems (OLSs). While OSaaS facilitates transparent connectivity and resource sharing among users, it raises concerns over potential network vulnerabilities due to shared fiber access and inter-channel interference, such as fiber nonlinearity and amplifier-based crosstalk. These challenges are exacerbated in multi-user environments, complicating the identification and localization of service interferences. To reduce system disruptions and system repair costs, it is beneficial to detect and identify such interferences in a timely manner. Addressing these challenges, this paper introduces a machine learning (ML)-based architecture for network operators to detect and attribute interferences to specific OSaaS users while being blind to the users’ internal spectrum details. Our methodology leverages available coarse power measurements and operator channel performance data, bypassing the need for internal user information of wide-band shared spectra. Experimental studies conducted on a 190 km optical line system in the Open Ireland testbed, with three OSaaS users, demonstrate the model’s capability to accurately classify the source of interferences, achieving a classification accuracy of 90.3%.

Problem

• Detect interference in shared optical spectrum without user details
• Identify interference sources in multi-user OSaaS environments
• Reduce network disruptions using ML-based interference classification

Innovation

• Machine Learning based interference detection
• Uses coarse power measurements
• Blind to user spectrum details