140 Gbps Photonics-Aided THz Wireless Communication Around 400 GHz Band Based on Artificial Neural Network Equalizer

Yuancheng Cai,Liyao Zhang,Jiao Zhang,Bingchang Hua,Mingzheng Lei,Yucong Zou,Guo Zhao,Jianjun Yu,Min Zhu
DOI: https://doi.org/10.1109/ACP55869.2022.10089032
2022-01-01
Abstract:We propose and experimentally verify a 400 GHz photonic-aided THz wireless communication system enabled by an artificial neural network (ANN) equalizer. The bit error rate (BER) performance is studied at different THz carrier frequencies from 370 GHz to 430 GHz. In addition, by means of using the piecewise learning rate approach, the ANN equalizer can improve the BER performance by approximate one order of magnitude, as compared with the traditional least mean square linear equalizer. We verify the BER performance at different baud rates from 32 GBd to 38 GBd in 390 GHz THz band. The results show that the piecewise learning rate enabled ANN equalizer can achieve a net rate of 140 Gbps after transmission over 20 km standard single mode fiber and 1 m wireless distances.
What problem does this paper attempt to address?