Gated Recurrent Unit based Autoencoder for Optical Link Fault Diagnosis in Passive Optical Networks
2021 EUROPEAN CONFERENCE ON OPTICAL COMMUNICATION (ECOC)(2021)
摘要
We propose a deep learning approach based on an autoencoder for identifying and localizing fiber faults in passive optical networks. The experimental results show that the proposed method detects faults with 97% accuracy, pinpoints them with an RMSE of 0.18 m and outperforms conventional techniques.
更多查看译文
AI 理解论文
溯源树
样例
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要