Study On Optical Communications, Volume. 51, Issue 2, 240040-01(2025)
Research Progress of Machine Learning Algorithms Applied in FSO Communication Systems
Figures & Tables(9)
Fig. 2. M-QAM FSO communication system model applying machine learning algorithm
Fig. 3. BER performance of ANN and FCNN algorithm applied in OOK-FSO communication system under Strong, Moderate and Weak turbulence intensity
Fig. 4. Data-driven channel model using machine learning algorithm in FSO communication system[26]
Fig. 5. Distributions of predicted data gained from three deep learning-based channel models and the related fitting PDFs[26]
Fig. 7. Performance comparison between general detection methods and machine learning algorithms[14]
Table 1. Output quality and response using different machine learning algorithms in FSO communication systems
View tableView in ArticleTable 1. Output quality and response using different machine learning algorithms in FSO communication systems
机器学习算法 系统输出质量和响应 DL 恢复原始数据，改善BER性能 DNN 补偿非线性失真，提高EVM性能 GAN 低复杂度实现信道建模，学习FSO信道的统计分布特征 GRU 信道预测准确率可达80%~90% AS-SPGD 加快收敛，补偿波前像差 CNN 算法迭代次数少，处理信号图像效率高
Table 2. Performance of applying machine learning algorithms in different tasks
View tableView in ArticleTable 2. Performance of applying machine learning algorithms in different tasks
应用领域 机器学习算法在FSO通信系统中的性能分析 信号检测 信号检测速度快、降低BER，提高系统稳定性 信道建模 以数据驱动获得准确的信道模型，动态调整发送功率和接收参数，有效预测RSSI 辅助光学补偿 实时反馈和动态调整发送光束的形状和相位，补偿大气湍流引起的光学畸变
Tools
Get Citation
Copy Citation Text
Hainan LIU, Yufeng SHAO, Anrong WANG, Yaodong ZHU, Linjie YANG, Chao CHEN, Wenchen LI, Wenguang HU. Research Progress of Machine Learning Algorithms Applied in FSO Communication Systems[J]. Study On Optical Communications, 2025, 51(2): 240040-01
Paper Information
Category:
Received: Apr. 2, 2024
Accepted: --
Published Online: May. 22, 2025
The Author Email: Yufeng SHAO (syufeng@163.com)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20