基于深度学习的物理层无线通信技术:机遇与挑战  

Deep learning based physical layer wireless communication techniques: opportunities and challenges

作　　者：桂冠[1] 王禹 黄浩 GUI Guan;WANG Yu;HUANG Hao(College of Telecommunications and Information Engineering,Nanjing University of Posts and Telecommunications,Nanjing 210003,China)

机构地区：[1]南京邮电大学通信与信息工程学院,江苏南京210003

出　　处：《通信学报》2019年第2期19-23,共5页Journal on Communications

摘　　要：对无线通信系统的高可靠性与超高容量需求促进了第五代移动通信(5G)的发展,然而,随着通信系统的日益复杂,现有的物理层无线通信技术难以满足这些高的性能需求。目前,深度学习被认为是处理物理层通信的有效工具之一,基于此,主要探讨了深度学习在物理层无线通信中的潜在应用,并且证明了其卓越性能。最后,提出几个可能发展的基于深度学习的物理层无线通信技术。The development of the fifth-generation wireless communications(5G)system is promoted by the high requirements of the high reliability and super-high network capacity.However,existing communication techniques are hard to achieve the high requirements due to the more and more complexity design in 5G system.Currently,deep learning is considered one of effective tools to handle the physical layer wireless communications.Several potential applications based on deep learning were reviewed,and their effectiveness were confirmed.Finally,several potential techniques in deep learning based physical layer wireless communications were pointed out.

关 键 词：物理层无线通信 深度学习 深度神经网络 调制模式识别 波束成形 

分 类 号：TM929.5[电气工程—电力电子与电力传动]