Impairment Identification for PAM-4 Transceivers and Links Using Machine Learning
2021 OPTICAL FIBER COMMUNICATIONS CONFERENCE AND EXPOSITION (OFC)(2021)
摘要
We demonstrate simultaneous TDECQ estimation and impairment identification using convolutional neural networks. Robust training with representative impairments yielded identification accuracy of 100% for TDECQ >2.6 dB, when considering limited bandwidth, signal compression and SNR degradation. (C) 2021 Stephen E. Ralph
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要