Research Article
Artificial Neural Network Assisted Mitigation of Cross-modulation Distortion in Microwave Photonics Link
@INPROCEEDINGS{10.1007/978-3-030-90196-7_42, author={Yihui Yin and Wanli Yang and Xu Yang and Yong Qin and Hongtao Zhu}, title={Artificial Neural Network Assisted Mitigation of Cross-modulation Distortion in Microwave Photonics Link}, proceedings={Artificial Intelligence for Communications and Networks. Third EAI International Conference, AICON 2021, Xining, China, October 23--24, 2021, Proceedings, Part I}, proceedings_a={AICON}, year={2021}, month={11}, keywords={Microwave photonics Cross modulation distortion Artificial neural network Genetic algorithm}, doi={10.1007/978-3-030-90196-7_42} }- Yihui Yin
Wanli Yang
Xu Yang
Yong Qin
Hongtao Zhu
Year: 2021
Artificial Neural Network Assisted Mitigation of Cross-modulation Distortion in Microwave Photonics Link
AICON
Springer
DOI: 10.1007/978-3-030-90196-7_42
Abstract
A multi-carrier down-conversion microwave photonics link (MDC-MWPL) is designed to deliver the broadband radio frequency (RF) signals with multiple frequency and down-conversion the RF signal to intermediate frequency (IF) signal, contributing to the wide bandwidth, low loss, strong immunity to electromagnetic interference property of microwave photonics link. However, the link performance is often degraded by the cross-modulation distortion (XMD). So, a kind of artificial neural network genetic algorithm (ANN-GA) distortion compensation technique is proposed to mitigate the XMD of the MDC-MWPL. The trained artificial neural network fits the input-to-output mapping of the link and predicts the link output. Taking the predicted output as the individual fitness value of the genetic algorithm, the optimal compensation factor γ is found. Taking advantage of the γ, the XMD is mitigated by extracted and reconstructed compensation signal, with a suppression ratio of −65 dB. Different from the traditional digital distortion compensation method, the proposed technique can realize distortion compensation for any kinds of links, which is not limited to a fixed microwave photonics link and its mathematical model, improving the intelligence and flexibility of microwave photonic link linearization design.
