[TuG3] Algorithms for Photonic Computing

2:45 PM - 3:00 PM JST(5:45 AM - 6:00 AM UTC)

[TuG3-4] A Novel CNN-Based Nonlinearity Correction Method for FMCW LiDAR

○Hong Ye¹, Yi Hao¹, Qingyang Zhu¹, Qian Li², H. Y. Fu¹ (1.Tsinghua University, 2.Peking University)

Keywords:

Deep learning for photonic device and applications,Digital signal processing,advanced modulation,coding and multiplexing

A novel CNN-based nonlinearity correction method for the FMCW LiDAR system is presented, which is data efficient and model-free. Effectiveness is experimentally validated through a 97.6% RMS reduction, decreasing from 215.65 GHz to 5.16 GHz.