Hybrid Security and Fault Detection in Short-Haul Fiber Links: Integrating Wave Equation Inversion with 5σ Backscatter Analysis

Fuente: Zenodo
Saved in:
Bibliographic Details
Main Author: Ben Hador, Ronen
Format: Recurso digital
Language:English
Published: Zenodo 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866901668337025024
author Ben Hador, Ronen
author_facet Ben Hador, Ronen
contents <p>Modern data center and metropolitan area networks (MANs) utilize dense, short-haul fiber<br>segments where traditional Optical Time-Domain Reflectometry (OTDR) is severely<br>limited by pulse overlap, inherent noise (CRN), and ambiguity from multiple reflections.<br>This ambiguity renders conventional fault detection and, critically, security monitoring,<br>unreliable. This work presents a novel, hybrid methodology to resolve these challenges,<br>delivering high-fidelity, autonomous fiber monitoring.The core of the solution is a<br>two-branch system. The first branch utilizes a Wave Equation Inversion (WEI) kernel to<br>model and invert the complex, overlapping backscatter signal, enabling the unambiguous,<br>high-resolution mapping of the fiber’s true reflectivity profile, κ(z). The second branch<br>introduces a rigorous statistical α(z) detection scheme that applies a 5σ threshold to the<br>localized backscatter decay parameter to identify sub-threshold attenuation events<br>characteristic of stealthy, non-reflective fiber taps.The entire architecture is integrated into<br>an autonomous system powered by Physics-Informed Neural Networks (PINNs) for<br>real-time data ingestion and processing, with results delivered to an LLM-based RAG<br>system for diagnostic and prescriptive maintenance output. This approach transforms<br>fiber monitoring from a signal processing problem into an interpretable, mathematically rigorous, and deployable security solution</p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_18007745
institution Zenodo
language eng
publishDate 2025
publisher Zenodo
record_format zenodo
spellingShingle Hybrid Security and Fault Detection in Short-Haul Fiber Links: Integrating Wave Equation Inversion with 5σ Backscatter Analysis
Ben Hador, Ronen
Fiber Optic Security
RBH Technologies
Wave Equation Inversion
Backscatter Analysis
Autonomous Monitoring
<p>Modern data center and metropolitan area networks (MANs) utilize dense, short-haul fiber<br>segments where traditional Optical Time-Domain Reflectometry (OTDR) is severely<br>limited by pulse overlap, inherent noise (CRN), and ambiguity from multiple reflections.<br>This ambiguity renders conventional fault detection and, critically, security monitoring,<br>unreliable. This work presents a novel, hybrid methodology to resolve these challenges,<br>delivering high-fidelity, autonomous fiber monitoring.The core of the solution is a<br>two-branch system. The first branch utilizes a Wave Equation Inversion (WEI) kernel to<br>model and invert the complex, overlapping backscatter signal, enabling the unambiguous,<br>high-resolution mapping of the fiber’s true reflectivity profile, κ(z). The second branch<br>introduces a rigorous statistical α(z) detection scheme that applies a 5σ threshold to the<br>localized backscatter decay parameter to identify sub-threshold attenuation events<br>characteristic of stealthy, non-reflective fiber taps.The entire architecture is integrated into<br>an autonomous system powered by Physics-Informed Neural Networks (PINNs) for<br>real-time data ingestion and processing, with results delivered to an LLM-based RAG<br>system for diagnostic and prescriptive maintenance output. This approach transforms<br>fiber monitoring from a signal processing problem into an interpretable, mathematically rigorous, and deployable security solution</p>
title Hybrid Security and Fault Detection in Short-Haul Fiber Links: Integrating Wave Equation Inversion with 5σ Backscatter Analysis
topic Fiber Optic Security
RBH Technologies
Wave Equation Inversion
Backscatter Analysis
Autonomous Monitoring
url https://doi.org/10.5281/zenodo.18007745