Paper
18 March 2024 Bidirectional demarcated vector-based bidirectional link failure localization method in optical network
Author Affiliations +
Proceedings Volume 13104, Advanced Fiber Laser Conference (AFL2023); 131040D (2024) https://doi.org/10.1117/12.3017098
Event: Advanced Fiber Laser Conference (AFL2023), 2023, Shenzhen, China
Abstract
In optical networks, link failures caused by fiber cuts occur frequently. Existing studies fails to take cost optimization, accuracy, and latency simultaneously when it comes to bidirectional link failure localization. In this paper, we propose a Bidirectional Demarcated Vector (BDV) algorithm, which is designed to guarantee 100% failure localization while minimizing costs. Concretely, the optical network topology is reconstructed as a digraph by using bidirectional demarcated vectors firstly. The direction of the directed edge between the same two nodes is opposite. Then an integer linear programming (ILP) model is designed to generate a directed monitoring trail based on the reconstructed optical network topology by adjusting the search space, constraint conditions, and link state adaptability. In addition, to maximize the utilization rate of link resources when selecting the optimal localization strategy, the bidirectional traffic factor is considered to compare the costs of different trajectories. Simulation results show that, the BDV method is about 20% lower than the Detection Selection algorithm in terms of localization cost, and is 50%-60% lower in latency than the Fusion algorithm (denoted by AA) and the Detection Selection algorithm (DS). Furthermore, the BDV algorithm surpasses the Random Walk-based failure localization algorithm (RWL) by about 10% in terms of localization accuracy.
(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Jiahui Wang, Haojun Zhang, and Zhipeng Zhao "Bidirectional demarcated vector-based bidirectional link failure localization method in optical network", Proc. SPIE 13104, Advanced Fiber Laser Conference (AFL2023), 131040D (18 March 2024); https://doi.org/10.1117/12.3017098
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Optical networks

Optical sensing

Failure analysis

Binary data

Detection and tracking algorithms

Algorithm development

Design

Back to Top