Optical Network Multi-fault Localization based on Network Topology and Knowledge Graph

HAN Jian-xing; LI Zhuo-tong; JING Yin-ji; ZHAO Yong-li; ZHANG Jie

doi:10.13756/j.gtxyj.2022.04.006

您当前的位置：

首页 >

文章列表页 >

Optical Network Multi-fault Localization based on Network Topology and Knowledge Graph

更新时间：2022-08-01

- Optical Network Multi-fault Localization based on Network Topology and Knowledge Graph
- Study on Optical Communications Issue 4, Pages: 27-30,51(2022)
- 作者机构：
  
  北京邮电　大学电子工程学院，北京　100876
- 作者简介：
- 基金信息：
- DOI：10.13756/j.gtxyj.2022.04.006
  CLC： TN915
- Published：10 August 2022，
  
  Received：24 November 2021，
扫描看全文
HAN Jian-xing, LI Zhuo-tong, JING Yin-ji, et al. Optical Network Multi-fault Localization based on Network Topology and Knowledge Graph. [J]. Study on Optical Communications (4):27-30,51(2022)
DOI：

HAN Jian-xing, LI Zhuo-tong, JING Yin-ji, et al. Optical Network Multi-fault Localization based on Network Topology and Knowledge Graph. [J]. Study on Optical Communications (4):27-30,51(2022) DOI： 10.13756/j.gtxyj.2022.04.006.

摘要

随着光网络结构越加庞大复杂，光网络故障更易发生。网络故障发生后，由于告警间的衍生特性，根源告警会衍生出多个衍生告警，因此网管系统会收到告警风暴。由于故障和告警间关系复杂，定位网络故障，尤其是多故障，难度也急剧攀升。针对此问题，文章将擅长管理海量信息并揭示数据规律的知识图谱技术引入到光网络中，告警知识图谱蕴含了丰富的告警间关系，结合人工智能技术图神经网络（GNN）后可完成知识引导的网络故障根因自动推理，符合运维人员运维过程中的故障定位思路。进一步地，在故障定位过程中增加了网络拓扑信息，提升了知识图谱的知识维度，解除了单故障场景的限制，并在多故障定位场景中得到了较高的准确率。

Abstract

As the optical network structure becomes larger and more complex

optical network faults are more likely to occur. After a network fault occurs

due to the derivative characteristics of network alarms

the root cause alarms will generate multiple derivative alarms. Therefore

after network faults occur

the network management system will receive alarm storms. Due to the complex relationship between faults and alarms

the difficulty of locating network faults

especially multiple faults

has also risen sharply. In response to this problem

the knowledge graph technology that is good at managing massive amounts of information and revealing the characteristics of data is introduced into optical networks. The alarm knowledge graph contains rich relationships between alarms

which can be completed by combining Graph Neural Network (GNN) technology. The knowledge-guided automatic reasoning of the root cause of network faults is in line with the fault location ideas of the operation and maintenance personnel in the operation and maintenance process. Further

the network topology information is added in the fault location process

and the knowledge dimension of the knowledge graph is improved. The limitation of the single-fault scenario is lifted

and a high accuracy rate is obtained in the multi-fault location scenario.

关键词

知识图谱光网络故障定位

Keywords

knowledge graphoptical networksfault localization

references

Singhal A. Introducing the Knowledge Graph:Things, not Strings[EB/OL].(2012-05-16)[2021-11-25]. https://blog.google/products/search/introducingknowledge-graph-things-not/https://blog.google/products/search/introducingknowledge-graph-things-not/.

Schlichtkrull M, Kipf T N, Bloem P, et al. Modeling Relational Data with Graph Convolutional Networks[C]//European Semantic Web Conference. Heraklion, Greece: Springer, 2018: 593-607.

Zhang M, Chen Y. Link Prediction based on Graph Neural Networks[J]. Advances in Neural Information Processing Systems, 2018, 31:5165-5175.

Wang Z, Chen T, Ren J, et al. Deep Reasoning with Knowledge Graph for Social Relationship Understanding[DB/OL]. (2018-07-02)[2021-11-25]. https://arxiv.org/abs/1807.00504https://arxiv.org/abs/1807.00504.

Gori M, Monfardini G, Scarselli F. A New Model for Learning in Graph Domains[C]//2005 IEEE International Joint Conference on Neural Networks. Montreal, QC, Canada: IEEE, 2005: 1555942.

Zhou J, Cui G, Zhang Z, et al. Graph Neural Networks:A Review of Methods and Applications[DB/OL]. (2021-10-6)[2021-11-25]. https://arxiv.org/abs/1812.08434https://arxiv.org/abs/1812.08434.

Li Y, Tarlow D, Brockschmidt H M, et al. Gated Graph Sequence Neural Networks[DB/OL]. (2017-09-22)[2021-11-25]. https://arxiv.org/abs/1511.05493https://arxiv.org/abs/1511.05493.

更多指标>

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Risk-aware Optical Network Service Restoration Algorithm under Persistent Disasters

Survivability Research on Optical Networks based on Quantum Key Distribution

Design the configurations of the OADMs based on AWGs

Related Author

No data

Related Institution

Electric Power Research Institude, State Grid Xinjiang Electric Co., Ltd.

State Key Laboratory of Information Photonics and Optical Communications, Beijing University of Posts and Telecommunications

Dispatching and Control Center of Electric Power Co., Ltd., State Grid Xinjiang Electric Co., Ltd.

State Grid Laboratory of Electric Power Communication Network Technology, Global Energy Interconnection Research Institute Co., Ltd.

State Key Laboratory of Information Photonics and Optical Communications, Beijing University of Posts and Telecommunications

⁰