毫米波MIMO的DNN混合预编码梯度优化方法

doi:10.19665/j.issn1001-2400.2022.01.021

西安电子科技大学学报 ›› 2022, Vol. 49 ›› Issue (1): 202-207.doi: 10.19665/j.issn1001-2400.2022.01.021

毫米波MIMO的DNN混合预编码梯度优化方法

王勇¹(),王喜媛²(),任泽洋³()

1.西安电子科技大学网络与信息安全学院,陕西西安 710071
2.西安电子科技大学信息科学研究中心,陕西西安 710071
3.北京邮电大学国际学院,北京 100876

收稿日期:2021-11-29 出版日期:2022-02-20 发布日期:2022-04-27
作者简介:王勇(1978—),男,副教授,博士,E-mail: wangyong@mail.xidian.edu.cn;|王喜媛(1979—),女,讲师,博士,E-mail: xywang1@mail.xidian.edu.cn;|任泽洋(2001—),男,北京邮电大学学生,E-mail: 6268779@qq.com
基金资助:
国家重点研发计划(2018YFB0804103)

Algorithm for gradient optimization of hybrid precoding based on DNN in the millimeter wave MIMO system

WANG Yong¹(),WANG Xiyuan²(),REN Zeyang³()

1. School of Cyber Engineering,Xidian University,Xi'an 710071,China
2. Information Science Research Center,Xidian University,Xi'an 710071,China
3. International School,Beijing University of Posts and Telecommunications,Beijing 100876,China

Received:2021-11-29 Online:2022-02-20 Published:2022-04-27

摘要/Abstract

摘要：

毫米波多输入多输出的混合预编码是降低硬件复杂度和能量消耗的重要方法。为降低优化处理的复杂性并提升频谱效率,提出了一种支持深度学习的混合预编码优化算法。为了消除子信道之间由信噪比差异导致部分子信道误码率较高进而对整体误码率产生的不良影响,通过基于块对角化的几何均值分解和深度神经网络的训练来选择混合预编码器,将预编码器的优化选择视为深度神经网络中的映射,以优化大规模多输入多输出的混合预编码过程。将频谱效率的优化问题近似归结为全数字预编码器和混合预编码器之间的欧氏距离的最小化问题,利用有限数量的射频链路实现吞吐量的改善。性能分析和仿真结果都表明,由于采用改进的梯度计算算法和单循环迭代结构,基于深度神经网络的方法能够最小化毫米波多输入多输出的误码率并提高频谱效率,同时显著地降低了所需的计算复杂度。当频谱效率为50 bit/(s· Hz)时,信噪比可节省3 dB。不同方案达到相同误码率时,信噪比可节省5 dB以上,并具有更好的稳健性。

关键词: 混合预编码, 频谱效率, 几何平均分解, 深度神经网络, 多输入多输出

Abstract:

Hybrid precoding of millimeter wave Multi-Input Multi-Output (MIMO) is an important method to reduce hardware complexity and energy consumption.In order to reduce the complexity of optimization processing and improve spectral efficiency,a hybrid precoding fast optimization algorithm based on deep learning is proposed.The difference in signal-to-noise ratio between subchannels may lead to a poor bit error rate performance.The hybrid precoder is selected by geometric mean decomposition (GMD) of block diagonalization and training based on the deep neural network (DNN).The optimal selection of the precoder is regarded as the mapping relationship in the DNN to optimize the hybrid precoding process of the large-scale MIMO.The optimization problem of spectral efficiency is approximately reduced to the minimization of the Euclidean distance between all digital precoders and hybrid precoders,and the throughput is improved by using a limited number of RF links.Performance analysis and simulation results show that due to the improved gradient algorithm and single cycle iterative structure,the DNN based method can minimize the bit error rate (BER) of the millimeter wave MIMO and improve the spectral efficiency,while significantly reducing the required computational complexity.When the spectral efficiency is 50bps/Hz,the SNR can be saved by 3dB.If different schemes achieve the same bit error rate,the SNR can be saved by more than 5dB and have better robustness.

Key words: hybrid precoding, spectral efficiency, geometric mean decomposition, deep neural network, multi-input multi-output

中图分类号:

TP391

王勇,王喜媛,任泽洋. 毫米波MIMO的DNN混合预编码梯度优化方法[J]. 西安电子科技大学学报, 2022, 49(1): 202-207.

WANG Yong,WANG Xiyuan,REN Zeyang. Algorithm for gradient optimization of hybrid precoding based on DNN in the millimeter wave MIMO system[J]. Journal of Xidian University, 2022, 49(1): 202-207.

图/表 5

参考文献 15

[1]	AYACH O E, RAJAGOPAL S, ABU-SURRA S, et al. Spatially Sparse Precoding in Millimeter Wave MIMO Systems[J]. IEEE Transactions on Wireless Communications, 2014, 13(3):1499-1513. doi: 10.1109/TWC.2014.011714.130846
[2]	YU X, SHEN J C, ZHANG J, et al. Alternating Minimization Algorithms for Hybrid Precoding in Millimeter Wave MIMO Systems[J]. IEEE Journal on Selected Areas in Communications, 2016, 10(3):485-500.
[3]	GAO X, DAI L, HAN S, et al. Energy-Efficient Hybrid Analog and Digital Precoding for mmwave MIMO Systems with Large Antenna Arrays[J]. IEEE Journal on Selected Areas in Communications, 2016, 34(4):998-1009. doi: 10.1109/JSAC.2016.2549418
[4]	ALKHATEEB A, LEUS G, HEATH R W. Limited Feedback Hybrid Precoding for Multi-User Millimeter Wave Systems[J]. IEEE Transactions on Wireless Communications, 2015, 14(11):6481-6494. doi: 10.1109/TWC.2015.2455980
[5]	CHEN C E, TSAI Y C, YANG C H. An Iterative Geometric Mean Decomposition Algorithm for MIMO Communications Systems[J]. IEEE Transactions on Wireless Communications, 2015, 14(1):343-352. doi: 10.1109/TWC.2014.2347051
[6]	JIN J, ZHENG Y R, CHEN W, et al. Hybrid Precoding for Millimeter Wave MIMO Systems:A Matrix Factorization Approach[J]. IEEE Transactions on Wireless Communications, 2018, 17(5):3327-3339. doi: 10.1109/TWC.2018.2810072
[7]	ZHANG E, HUANG C. On Achieving Optimal Rate of Digital Precoder by RF-baseband Codesign for MIMO Systems[C]// Proceedings of IEEE Vehicular Technology Conferrence.Piscataway:IEEE, 2014:1-5.
[8]	WANG G, ASCHEID G. Joint Pre/Post-processing Design for Large Millimeter Wave Hybrid Spatial Processing Systems[C]// Proceedings of 20th European Wireless Conference. Barcelona:VDE, 2014:1-6.
[9]	ZHANG Y, DONG X, ZHANG Z. Machine Learning-Based Hybrid Precoding with Low-Resolution Analog Phase Shifters[J/OL]. [2021-11-10].DOI: 10.1109/LCOMM,2020,3025025. doi: 10.1109/LCOMM,2020,3025025
[10]	MAO B. Routing or Computing? The Paradigm Shift Towards Intelligent Computer Network Packet Transmission Based on Deep Learning[J]. IEEE Transactions on Computer, 2017, 66(11):1946-1960. doi: 10.1109/TC.2017.2709742
[11]	GUI G, HUANG H, SONG Y, et al. Deep Learning for an Effective Non-Orthogonal Multiple Access Scheme[J]. IEEE Transactions on Vehicular Technology, 2018, 67(9):8440-8450. doi: 10.1109/TVT.2018.2848294
[12]	HUANG H, YANG J, SONG Y, et al. Deep Learning for Super-Resolution Channel Estimation and DOA Estimation based Massive MIMO System[J]. IEEE Transactions on Vehicular Technology, 2018, 67(9):8549-8560. doi: 10.1109/TVT.2018.2851783
[13]	HUANG H, GUI G, SARI H, et al. Deep Learning for Super-Resolution DOA Estimation in Massive MIMO Systems[C]// Proceedings of 2018 IEEE 88th Vehicular Technology Conference (VTC Fall).Piscataway:IEEE, 2018.
[14]	YAN L, HAN C, YUAN J. A Dynamic Array-of-Subarrays Architecture and Hybrid Precoding Algorithms for Terahertz Wireless Communications[J]. IEEE Journal on Selected Areas in Communications, 2020, 38:2041-2056. doi: 10.1109/JSAC.2020.3000876
[15]	耿烜, 孙作雷, 刘锋, 等. 块对角化的最小均方误差几何均值分解矢量预编码[J]. 西安电子科技大学学报, 2013, 40(2):207-212.
	GENG Xuan, SUN Zuolei, LIU Feng, et al. Block Diagonalized MMSE GMD Vector Precoding[J]. Journal of Xidian University, 2013, 40(2):207-212.

算法	复杂度
文中算法	O(×N_t×N_s)
交替最小算法^[9]	O(××)
SVD算法^[11]	O(2N_rN_s((K+N_rN_s)))

毫米波MIMO的DNN混合预编码梯度优化方法

Algorithm for gradient optimization of hybrid precoding based on DNN in the millimeter wave MIMO system

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 5

参考文献 15

相关文章 15

Metrics

本文评价

推荐阅读 0

[1]	刘敏提, 曾操, 胡树林, 陈建忠, 李军, 李世东, 廖桂生. 存在幅相误差时二维稳健超分辨测角算法[J]. 西安电子科技大学学报, 2024, 51(3): 55-62.
[2]	赵浩钦, 杨政, 司江勃, 石嘉, 严少虎, 段国栋. 一种聚类辅助的智能频谱分配技术研究[J]. 西安电子科技大学学报, 2023, 50(6): 1-12.
[3]	史加荣,李金红. 新型深度矩阵分解及其在推荐系统中的应用[J]. 西安电子科技大学学报, 2022, 49(3): 171-182.
[4]	蒋芳,黄兴,胡梦钰,王翊,许耀华,胡艳军. 降噪自编码器辅助的下行MIMO-SCMA编解码方法[J]. 西安电子科技大学学报, 2022, 49(3): 74-82.
[5]	曹鹏程,刘伟伟,刘光杰,茅伟伟,戴跃伟. 一种基于预编码的MIMO无线隐信道[J]. 西安电子科技大学学报, 2022, 49(2): 42-49.
[6]	张书伟,李俊民. 一种复杂监控场景下的人体检测算法[J]. 西安电子科技大学学报, 2021, 48(5): 68-77.
[7]	曹艺,蔡晓东. 一种有效的困难样本学习策略[J]. 西安电子科技大学学报, 2021, 48(3): 99-105.
[8]	刘洁怡,公茂果,詹涛,李豪,张明阳. 一种深度神经网络多站雷达系统干扰鉴别方法[J]. 西安电子科技大学学报, 2021, 48(2): 133-138.
[9]	张华,高浩然,杨兴国,李文敏,高飞,温巧燕. TargetedFool:一种实现有目标攻击的算法[J]. 西安电子科技大学学报, 2021, 48(1): 149-159.
[10]	张树栋,高海昌,曹曦文,康帅. 针对ASR系统的快速有目标自适应对抗攻击[J]. 西安电子科技大学学报, 2021, 48(1): 168-175.
[11]	常新旭,张杨,杨林,寇金桥,王昕,徐冬冬. 融合多头自注意力机制的语音增强方法[J]. 西安电子科技大学学报, 2020, 47(1): 104-110.
[12]	袁文浩,梁春燕,娄迎曦,房超,王志强. 一种时频平滑的深度神经网络语音增强方法[J]. 西安电子科技大学学报, 2019, 46(4): 130-136.
[13]	乔瑞秀,陈刚,龚国良,鲁华祥. 一种高性能可重构深度卷积神经网络加速器[J]. 西安电子科技大学学报, 2019, 46(3): 130-139.
[14]	陈晓范,申海杰,边倩,王振铎,田新志. 结合注意力机制的人脸超分辨率重建[J]. 西安电子科技大学学报, 2019, 46(3): 148-153.
[15]	袁文浩,娄迎曦,梁春燕,王志强. 感知联合优化的深度神经网络语音增强方法[J]. 西安电子科技大学学报, 2019, 46(2): 89-94.