Chinese Optics Letters, Volume. 22, Issue 7, 070602(2024)
Transfer-learning multi-input multi-output equalizer for mode-division multiplexing systems
Fig. 4. BER results of DT versus (a) Trecycling and (b) Ldata with TL-MIMO, while the data of DT and DS are collected at different time moments.
Fig. 5. BER versus μ-K combination in (a) DS and (b) DT with the data collected at different time moments.
Fig. 6. BER versus Trecycling × Ldata, illustrating (a) without and (b) with the full TL-MIMO scheme.
Fig. 7. Statistical BER results for 30 datasets in DT comparing different MIMO schemes with respect to (a) Trecycling and (b) Ldata.
Fig. 8. In L-H case of launched power, BER results versus (a) Trecycling and (b) Ldata; in H-L case of launched power, BER results versus (c) Trecycling and (d) Ldata for 30 groups of data in DT.
Fig. 10. In L-H OSNR case, BER results versus (a) Trecycling and (b) Ldata; in H-L OSNR case, BER results versus (c) Trecycling and (d) Ldata for 30 groups of data in DT.
Get Citation
Copy Citation Text
Tianfeng Zhao, Feng Wen, Mingming Tan, Baojian Wu, Bo Xu, Kun Qiu, "Transfer-learning multi-input multi-output equalizer for mode-division multiplexing systems," Chin. Opt. Lett. 22, 070602 (2024)
Category: Fiber Optics and Optical Communications
Received: Mar. 11, 2024
Accepted: Mar. 29, 2024
Published Online: Aug. 2, 2024
The Author Email: Feng Wen (fengwen@uestc.edu.cn)