[ThG1] AI-Based Optical Network Management

10:30 AM - 10:45 AM JST(1:30 AM - 1:45 AM UTC)

[ThG1-5] Concurrent Fault Prediction based Multi-View Feature Embedding for Optical Transport Networks

○Weijie Yang¹, Chunyu Zhang¹, Min Zhang¹, Xunjie Jiang², Yanlin Fan², Zhongbo Bi², Jiansheng Xiong², Danshi Wang¹ (1. State Key Laboratory of Information Photonics and Optical Communications, Beijing University of Posts and Telecommunications (China), 2. The Intelligent Network Innovation Center of Chinaunicom (China))

Keywords:

Optical core/metro network reliability and protection & restoration,Optical core/metro network architecture,design,control,and management,Core/Metro Network and Subsystem,Optical network security management

We propose a concurrent multi-fault prediction method that effectively embeds multi-view fault features for complex OTNs. This method can concurrently predict 23 types of faults with a zero false negative rate in real-world scenarios.