基于数据增强技术的光传送网健康度模型测评研究

王郁; 唐都; 李芳; 徐云斌

doi:10.11959/j.issn.1000-0801.DXKX250673

您当前的位置：

首页 >

文章列表页 >

基于数据增强技术的光传送网健康度模型测评研究

研究与开发 | 更新时间：2026-07-21

- 基于数据增强技术的光传送网健康度模型测评研究
- Research on evaluation of optical transport network health model based on data augmentation technology
- 电信科学 2026年42卷第6期页码：158-173
- 作者机构：
  
  中国信息通信研究院，北京 100091
- 作者简介：
  
  王郁（1975- ），女，中国信息通信研究院正高级工程师，主要研究方向为光网络智能管控、自智光网络、光网络数字孪生、人工智能与光网络融合。
  唐都（1998- ），男，博士，中国信息通信研究院工程师，主要研究方向为高速光通信、光网络通感一体、卫星光通信、光网络智能化。
  李芳（1974- ），女，中国信息通信研究院正高级工程师，主要研究方向为高速光通信、分组传送、4G/5G 移动承载和光网络智能化。
  徐云斌（1978- ），男，博士，中国信息通信研究院正高级工程师，主要从事智能光网络新技术研究、标准研制、测试验证等工作。
- 基金信息：
  
  工业和信息化部制造业专项“2024年光网络新技术验证项目”(20240248)
- DOI：10.11959/j.issn.1000-0801.DXKX250673
  中图分类号： TN929.11
- 收稿：2025-11-20，
  
  修回：2025-12-22，
  
  录用：2025-12-29，
  
  网络首发：2026-07-21，
  
  纸质出版：2026-06-20
- 稿件说明：
移动端阅览
王郁,唐都,李芳等.基于数据增强技术的光传送网健康度模型测评研究[J].电信科学,2026,42(06):158-173.

Wang Yu,Tang Du,Li Fang,et al.Research on evaluation of optical transport network health model based on data augmentation technology[J].Telecommunications Science,2026,42(06):158-173.
王郁,唐都,李芳等.基于数据增强技术的光传送网健康度模型测评研究[J].电信科学,2026,42(06):158-173. DOI： 10.11959/j.issn.1000-0801.DXKX250673.

Wang Yu,Tang Du,Li Fang,et al.Research on evaluation of optical transport network health model based on data augmentation technology[J].Telecommunications Science,2026,42(06):158-173. DOI： 10.11959/j.issn.1000-0801.DXKX250673.

摘要

随着光传送网（optical transport network，OTN）网络向高阶自智生态演进，OTN健康度模型的预测准确度直接关系到网络优化决策的可靠性。为突破传统通用时间序列增强方法缺乏物理约束的局限，提出一种基于数据增强技术与OTN物理传输特征相结合的健康度模型准确度测评方法。该方法基于OTN关键性能参数的物理变化规律设计基底函数，结合函数发生器与光网络仿真器，构建“物理引导—数学变换—光域验证”的数据增强机制，生成高保真合成数据集，模拟渐变、突变等多种性能变化场景，形成用于模型训练与验证的基准数据集。此外，详细设计了包含数据加载、增强、处理、模型训练与测评的完整流程，并通过实验验证了该方法的有效性。

Abstract

As optical transport network (OTN) evolve toward a higher-level autonomous and intelligent ecosystem

the prediction accuracy of OTN health models was directly related to the reliability of network optimization decisions. To overcome the limitation of conventional general time-series augmentation methods lacking physical constraints

an evaluation method for health model accuracy was proposed by integrating data augmentation technology with OTN physical transmission characteristics. In this method

basis functions were designed based on the physical variation patterns of key OTN performance parameters

and a data augmentation mechanism of “physics-guided mathematical transformation followed by optical-domain verification” was constructed by combining a function generator with an optical network simulator. High-fidelity synthetic datasets were generated to simulate various performance variation scenarios such as gradual changes and abrupt changes

thereby forming benchmark datasets for model training and validation. Furthermore

a complete workflow encompassing data loading

augmentation

processing

model training

and evaluation was detailed

and the effectiveness of the method was validated through experiments.

关键词

Keywords

references

TMF . Autonomous networks-business requirements & architecture: IG1218 [R ] . 2020 .

TMF . Autonomous networks-technical architecture: IG1230 [R ] . 2021 .

Zhang W Q , Chu P M , Huang K S , et al . Driving data generation using affinity propagation, data augmentation, and convolutional neural network in communication system [J ] . Internatlonal Journal of communication systems , 2021 , 34 ( 2 ): e3982 .

Liao C C , Zhao S Y , Wang X C , et al . EEG data augmentation method based on the Gaussian mixture model [J ] . Mathematics , 2025 , 13 ( 5 ): 729 .

Chen Z Y , Huang H Z , Deng Z W , et al . Shrinkage mamba relation network with out-of-distribution data augmentation for rotating machinery fault detection and localization under zero-faulty data [J ] . Mechanical Systems and Signal Processing , 2025 , 224 : 112145 .

Cai Z R , Ma W X , Wang X R , et al . The performance analysis of time series data augmentation technology for small sample communication device recognition [J ] . IEEE Transactions on Reliability , 2023 , 72 ( 2 ): 574 - 585 .

Mohammad R , Saeed F , Almazroi A A , et al . Enhancing intrusion detection systems using a deep learning and data augmentation approach [J ] . Systems , 2024 , 12 ( 3 ): 79 .

Liu Y C , Liu Y J , Song S , et al . GAN-based data augmentation for AI-enabled ATP in free space optical communication [J ] . IEEE Communications Letters , 2024 , 28 ( 5 ): 1067 - 1071 .

Yaméogo B L M , Charlton D W , Doucet D , et al . Trends in optical span loss detected using the time series decomposition method [J ] . Journal of Lightwave Technology , 2020 , 38 ( 18 ): 5026 - 5035 .

Amanu A A . Macro bending losses in single mode step index fiber [J ] . Advances in Applied Sciences , 2016 , 1 ( 1 ): 1 - 6 .

McDowell E J , Cui X Q , Yaqoob Z , et al . A generalized noise variance analysis model and its application to the characterization of 1/f noise [J ] . Optics Express , 2007 , 15 ( 7 ): 3833 - 3848 .

Dong J , Huang J C , Li T , et al . Observation of fundamental thermal noise in optical fibers down to infrasonic frequencies [J ] . Applied Physics Letters , 2016 , 108 ( 2 ): 021108 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于时间卷积网络的无监督入侵检测模型

基于改进长短期记忆网络的新能源场站网络安全评估方法研究

基于异构数据的电力短期负荷大数据预测方案

光纤通信系统技术的发展、挑战与机遇

分组增强型OTN设备实现及组网研究 *