深度学习赋能的水下无线光通信研究进展OA
Advances in deep learning-enabled underwater wireless optical communication
水下无线光通信(UWOC)兼具宽带宽、高数据传输速率以及低成本小体积发射器等优势,是海洋通信领域的重要支撑技术.然而水下信道环境复杂,受到吸收、散射、湍流等多种因素影响,传统方法在复杂动态场景中性能受限.近年来,深度学习(DL)因凭借强大的非线性拟合能力和数据驱动优势,为UWOC的研发提供了重要支撑.总结了DL在UWOC中信道估计、信号检测、调制格式识别、轨道角动量(OAM)模态识别、组网优化等关键技术方向,最后对DL在未来UWOC中的应用进行了总结和展望.
Underwater wireless optical communication(UWOC)offers advantages such as wide bandwidth,high data transmis-sion rates,and low-cost,compact transmitters,making it a key supporting technology in the field of marine communication.However,the underwater channel environment is complex and affected by various factors including absorption,scattering,and turbulence,which limit the performance of traditional methods in complex dynamic scenarios.In recent years,deep learning(DL),with its powerful nonlinear fitting capabilities and data-driven advantages,has provided important support for the devel-opment of UWOC.This paper summarizes the application of DL in key technical areas of UWOC,including channel estima-tion,signal detection,modulation format recognition,orbital angular momentum(OAM)mode recognition,and network optimi-zation.Finally,the future application of DL in UWOC is summarized and prospected.
朱宏娜;刘仲涛;朱宇轩;于水;郎子乂;罗斌;吴宗玲
西南交通大学 物理科学与技术学院,成都 610031西南交通大学 物理科学与技术学院,成都 610031西南交通大学 物理科学与技术学院,成都 610031西南交通大学 物理科学与技术学院,成都 610031西南交通大学 信息科学与技术学院,成都 610031西南交通大学 信息科学与技术学院,成都 610031西南交通大学 信息科学与技术学院,成都 610031
信息技术与安全科学
水下无线光通信深度学习信道估计信号检测调制格式识别轨道角动量模态识别组网优化
underwater wireless optical communicationdeep learningchannel estimationsignal detectionmodulation for-mat recognitionmodal identification of orbital angular momentumnetworking optimization
《光通信技术》 2026 (3)
1-11,11
国家重点研发计划项目(2021YFC3101402)资助中央高校基本科研业务费专项(2682025XJ020)资助.
评论