基于二维卷积的端到端无线通信系统研究
Research on End-to-End Wireless Communication System Based on Two-Dimensional Convolution
- 电子学报 2023年51卷第7期 页码：1725-1733
- 作者机构：
1.重庆邮电大学通信与信息工程学院,重庆 400065
2.先进网络与智能互联技术重庆市高校重点实验室,重庆 400065
3.泛在感知与互联重庆市重点实验室,重庆 400065
- 作者简介：
[ "刘乔寿 男，1979年8月出生于云南省曲靖市.现为重庆邮电大学副教授、硕士生导师.主要研究方向为5G超密集网络干扰协调、云边协同智能计算、FPGA智能算法加速、物联网系统及终端设备开发.E-mail: liuqs@cqupt.edu.cn" ]
[ "刘俊杰 男，1998年3月出生于湖北省天门市.目前就读于重庆邮电大学.主要研究方向为基于神经网络的端到端无线通信. E-mail: 15871764726@163.com" ]
[ "蔚淦丞 男，1998年6月出生于山西省朔州市.目前就读于重庆邮电大学.主要研究方向为自动编码器、基于深度学习的信道估计. E-mail: 758125724@qq.com" ]
周雄 男，1999年10月出生于湖北省洪湖市.目前就读于重庆邮电大学.主要研究方向为基于强化学习的信道估计. E-mail: zhouxiongz@163.com
- 基金信息：重庆市自然科学基金重点项目(cstc2020jcyj-zdxmX0024);重庆市研究生科研创新项目(CYS21297)
DOI：10.12263/DZXB.20220374
中图分类号： TN911.7收稿：2022-04-08，
修回：2023-02-16，
纸质出版：2023-07-25
- 稿件说明：
移动端阅览
刘乔寿,刘俊杰,蔚淦丞等.基于二维卷积的端到端无线通信系统研究[J].电子学报,2023,51(07):1725-1733.
LIU Qiao-shou,LIU Jun-jie,YU Gan-cheng,et al.Research on End-to-End Wireless Communication System Based on Two-Dimensional Convolution[J].ACTA ELECTRONICA SINICA,2023,51(07):1725-1733.
刘乔寿,刘俊杰,蔚淦丞等.基于二维卷积的端到端无线通信系统研究[J].电子学报,2023,51(07):1725-1733. DOI： 10.12263/DZXB.20220374.
LIU Qiao-shou,LIU Jun-jie,YU Gan-cheng,et al.Research on End-to-End Wireless Communication System Based on Two-Dimensional Convolution[J].ACTA ELECTRONICA SINICA,2023,51(07):1725-1733. DOI： 10.12263/DZXB.20220374.
摘要
针对现有端到端自动编译码器无线通信系统在多径信道中表现不佳的问题，本文提出了一种改进的自动编译码器端到端无线通信系统.在设计中，通过改变卷积核的尺寸，利用二维卷积来对抗多径效应引起的频率选择性衰落，并将传统OFDM（Orthogonal Frequency Division Multiplexing）模块同自动编译码器相结合，以此来增加系统应对多径信道的能力.同时，通过仿真分析一维卷积自动编译码器无线通信系统、传统OFDM无线通信系统以及本文改进的自动编译码器无线通信系统在多径信道下的性能.结果表明，在5径瑞利信道下并且以误块率（Block Error Rate，BLER）作为性能指标时，所提出的基于二维卷积的自动编译码器无线通信系统在64QAM调制下相比经典自动编译码器无线通信系统和传统OFDM无线通信系统分别提升了17%和60%的性能，本文的仿真分析给出了详细的对比说明.另外本文还分析了不同调制、不同信道时卷积核数量对系统性能的影响.
Abstract
Aiming at the problem that the existing end-to-end automatic codec (auto-codec) wireless communication systems do not perform well in multipath channels
this paper proposes an improved auto-codec end-to-end wireless communication system. In the design
by changing the size of the convolution kernel
the two-dimensional convolution is used to combat the frequency selective fading caused by the multipath effect
and the traditional OFDM (Orthogonal Frequency Division Multiplexing) module is combined with the auto-codec to increase the system's ability to cope with multipath. At the same time
the performance of one-dimensional convolutional auto-codec wireless communication system
traditional OFDM wireless communication system and the improved auto-codec wireless communication system under multipath channel are analyzed by simulation. The results show that the proposed auto-codec wireless communication system based on two-dimensional convolution
compared to the classical auto-codec wireless communication system and the traditional OFDM wireless communication system
improves 17% and 60% in terms of block error rate (BLER) respectively
under 5-path Rayleigh channel
64QAM modulation. The details are given in the simulation analysis. In addition
this paper also analyzes the impact of the number of convolution kernels on the system performance under different modulations and different channels.
关键词
Keywords
references
ZEHAVI E . 8-PSK trellis codes for a Rayleigh channel [J]. IEEE Transactions on Communications , 1992 , 40 ( 5 ): 873 - 884 .
LECUN Y , BENGIO Y , HINTON G . Deep learning [J]. Nature , 2015 , 521 ( 7553 ): 436 - 444 .
FELIX A , CAMMERER S , DÖRNER S , et al . OFDM-autoencoder for end-to-end learning of communications systems [C]// 2018 IEEE 19th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC) . Piscataway : IEEE , 2018 : 1 - 5 .
HE H T , WEN C K , JIN S , et al . Deep learning-based channel estimation for beamspace mmWave massive MIMO systems [J]. IEEE Wireless Communications Letters , 2018 , 7 ( 5 ): 852 - 855 .
NEUMANN D , WIESE T , UTSCHICK W . Learning the MMSE channel estimator [J]. IEEE Transactions on Signal Processing , 2018 , 66 ( 11 ): 2905 - 2917 .
YE H , LIANG L , LI G Y , et al . Deep learning-based end-to-end wireless communication systems with conditional GANs as unknown channels [J]. IEEE Transactions on Wireless Communications , 2020 , 19 ( 5 ): 3133 - 3143 .
SAMUEL N , DISKIN T , WIESEL A . Deep MIMO detection [C]// 2017 IEEE 18th International Workshop on Signal Processing Advances in Wireless Communications (SPAWC) . Piscataway : IEEE , 2017 : 1 - 5 .
YE H , LI G Y , JUANG B H . Power of deep learning for channel estimation and signal detection in OFDM systems [J]. IEEE Wireless Communications Letters , 2017 , 7 ( 1 ): 114 - 117 .
HUANG H J , GUO S , GUI G , et al . Deep learning for physical-layer 5G wireless techniques: Opportunities, challenges and solutions [J]. IEEE Wireless Communications , 2020 , 27 ( 1 ): 214 - 222 .
WANG Y , LIU M , YANG J , et al . Data-driven deep learning for automatic modulation recognition in cognitive radios [J]. IEEE Transactions on Vehicular Technology , 2019 , 68 ( 4 ): 4074 - 4077 .
NACHMANI E , MARCIANO E , BURSHTEIN D , et al . RNN decoding of linear block codes [EB/OL]. ( 2017 )[2022]. https://api.semanticscholar.org/CorpusID:18262730 https://api.semanticscholar.org/CorpusID:18262730 .
O’SHEA T , HOYDIS J . An introduction to deep learning for the physical layer [J]. IEEE Transactions on Cognitive Communications and Networking , 2017 , 3 ( 4 ): 563 - 575 .
WU N , WANG X D , LIN B , et al . A CNN-based end-to-end learning framework toward intelligent communication systems [J]. IEEE Access , 2019 , 7 : 110197 - 110204 .
PATZOLD M , SZCZEPANSKI A , YOUSSEF N . Methods for modeling of specified and measured multipath power-delay profiles [J]. IEEE Transactions on Vehicular Technology , 2002 , 51 ( 5 ): 978 - 988 .
European Telecommunications Standards Institute . 3GPP:TS25.105 [S/OL]. [ 2022-04-08 ]. http://www.3gpp.org http://www.3gpp.org .
0
浏览量
17
下载量
0
CSCD
- 网络PDF下载
- Meta-XML下载
- BibTeX下载
- Endnote下载
- RefMan 下载
- NoteExpress下载
- NoteFirst下载
关联资源
相关文章
相关作者
相关机构
- 技术支持由北京北大方正电子有限公司提供 京ICP备09064830号-19
京公网安备11010802024621 - 本系统建议在Chrome、 IE9+ 以上版本浏览器阅读本站内容，360浏览器请切换至极速模式
- Cookies帮助我们提供服务并提供个性化体验。使用本网站，即表示您同意我们使用Cookies