基于图神经网络的联合用户调度与波束成形优化算法

doi:10.11959/j.issn.1000-436x.2022133

通信学报 ›› 2022, Vol. 43 ›› Issue (7): 73-84.doi: 10.11959/j.issn.1000-436x.2022133

基于图神经网络的联合用户调度与波束成形优化算法

何世文¹^,²^,³, 袁军¹, 安振宇³, 张敏⁴, 黄永明²^,³, 张尧学⁵

¹ 中南大学计算机学院，湖南长沙 410083
² 东南大学移动通信国家重点实验室，江苏南京 210096
³ 紫金山实验室，江苏南京 211111
⁴ 湖南邮电职业技术学院信息通信学院，湖南长沙 410015
⁵ 清华大学计算机科学与技术系，北京 100084

修回日期:2022-05-30 出版日期:2022-07-25 发布日期:2022-06-01
作者简介:何世文（1978- ），男，湖南汝城人，博士，中南大学教授、博士生导师，主要研究方向为无线通信与网络、分布式学习与优化计算理论、智能物联网（AIoT）和大数据分析的基础理论研究与无线通信网络平台开发及先进理论技术验证
袁军（1997- ），男，安徽六安人，中南大学硕士生，主要研究方向为图神经网络理论及其应用
安振宇（1988- ），男，安徽蚌埠人，博士，网络通信与安全紫金山实验室高级工程师，主要研究方向为超可靠低时延通信、跨层优化、智能优化等
张敏（1974- ），女，湖南平江人，湖南邮电职业技术学院教授，主要研究方向为多用户通信、协作通信、绿色通信、大规模多输入多输出通信
黄永明（1977- ），男，江苏吴江人，东南大学教授、博士生导师，主要研究方向为MIMO无线通信、协作无线通信、微波无线通信及应用
张尧学（1956- ），男，湖南常德人，中国工程院院士，清华大学教授、博士生导师，主要研究方向为计算机网络、操作系统以及普适计算
基金资助:
国家自然科学基金资助项目(62171474);国家自然科学基金资助项目(61720106003);东南大学移动通信国家重点实验室开放研究基金资助项目(2022D03);OPPO广东移动通信有限公司研究基金资助项目(CN05202112160224)

GNN-based optimization algorithm for joint user scheduling and beamforming

Shiwen HE¹^,²^,³, Jun YUAN¹, Zhenyu AN³, Min ZHANG⁴, Yongming HUANG²^,³, Yaoxue ZHANG⁵

¹ School of Computer Science and Engineering, Central South University, Changsha 410083, China
² National Mobile Communications Research Laboratory, Southeast University, Nanjing 210096, China
³ Purple Mountain Laboratories, Nanjing 211111, China
⁴ School of Information and Communication, Hunan Post and Telecommunication College, Changsha 410015, China
⁵ Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China

Revised:2022-05-30 Online:2022-07-25 Published:2022-06-01
Supported by:
The National Natural Science Foundation of China(62171474);The National Natural Science Foundation of China(61720106003);The Open Research Fund of National Mobile Communications Research Laboratory of Southeast University(2022D03);OPPO Guangdong Mobile Com-munication Co., Ltd.Research Fund(CN05202112160224)

摘要/Abstract

摘要：

协作多点（CoMP）传输技术具有降低同频干扰和提高频谱效率的特点。对于 CoMP，用户调度与波束成形是2个分别位于媒体访问接入层和物理层的基本研究问题。在考虑用户服务质量需求下，重点研究用户调度与波束成形的联合优化问题，并以网络吞吐量最大化为目标。为了克服传统优化算法计算开销大且未有效利用网络历史数据信息的问题，提出了一种基于图神经网络联合用户调度与功率分配模型，并结合波束向量的解析公式，以实现联合用户调度与波束成形优化。仿真分析表明，所提算法能够以较低的计算开销实现与传统优化算法相匹配，甚至更优的性能表现。

关键词: 跨层优化, 图神经网络, 协作多点, 用户调度, 波束成形

Abstract:

The coordinated multi-point (CoMP) transmission technology has the characteristics of reducing co-channel interference and improving spectral efficiency.For the CoMP technology, user scheduling (US) and beamforming (BF) design are two fundamental research problems located in the media access control layer and the physical layer, respectively.Under the consideration of user service quality requirements, the joint user US-BF optimization problem was investigated with the goal of maximizing network throughput.To overcome the problem that the traditional optimization algorithm had high computational cost and couldn’t effectively utilize the network historical data information, a joint US and power allocation (M-JEEPON) model based on graph neural network was proposed to realize joint US-BF optimization by combining the beam vector analytical solution.The simulation results show that the proposed algorithm can achieve the performance matching or even better than traditional optimization algorithms with lower computational overhead.

Key words: cross-layer optimization, graph neural network, coordinated multi-point, user scheduling, beamforming

中图分类号:

TN92

何世文, 袁军, 安振宇, 张敏, 黄永明, 张尧学. 基于图神经网络的联合用户调度与波束成形优化算法[J]. 通信学报, 2022, 43(7): 73-84.

Shiwen HE, Jun YUAN, Zhenyu AN, Min ZHANG, Yongming HUANG, Yaoxue ZHANG. GNN-based optimization algorithm for joint user scheduling and beamforming[J]. Journal on Communications, 2022, 43(7): 73-84.

图/表 12

图1

图2

图3

图4

表1

图5

表2

表3

图6

图7

表4

图8

参考文献 26

[1]	BHUSHAN N , LI J Y , MALLADI D ,et al. Network densification:the dominant theme for wireless evolution into 5G[J]. IEEE Communications Magazine, 2014,52(2): 82-89.
[2]	BASSOY S , FAROOQ H , IMRAN M A ,et al. Coordinated multi-point clustering schemes:a survey[J]. IEEE Communications Surveys ＆ Tutorials, 2017,19(2): 743-764.
[3]	3GPP. Coordinated multi-point operation for LTE physical layer aspects,v11.2.0[R]. TR 36.19 R11, 2013.
[4]	SOLAIJA M S J , SALMAN H , KIHERO A B ,et al. Generalized coordinated multipoint framework for 5G and beyond[J]. IEEE Access, 2020,9: 72499-72515.
[5]	QAMAR F , DIMYATI K B , HINDIA M N ,et al. A comprehensive review on coordinated multi-point operation for LTE-A[J]. Computer Networks, 2017,123: 19-37.
[6]	MUHAMMED A J , MA Z , DING Z G ,et al. Resource allocation for energy-efficient NOMA system in coordinated multi-point networks[J]. IEEE Transactions on Vehicular Technology, 2021,70(2): 1577-1591.
[7]	LI M , COLLINGS I B , HANLY S V ,et al. Multicell coordinated scheduling with multiuser zero-forcing beamforming[J]. IEEE Transactions on Wireless Communications, 2016,15(2): 827-842.
[8]	ZHAI D S , ZHANG R N , CAI L ,et al. Energy-efficient user scheduling and power allocation for NOMA-based wireless networks with massive IoT devices[J]. IEEE Internet of Things Journal, 2018,5(3): 1857-1868.
[9]	LI Z D , CHEN W , WU Q Q ,et al. Robust beamforming design and time allocation for IRS-assisted wireless powered communication networks[J]. IEEE Transactions on Communications, 2022,70(4): 2838-2852.
[10]	FU B , XIAO Y , DENG H M ,et al. A survey of cross-layer designs in wireless networks[J]. IEEE Communications Surveys ＆ Tutorials, 2014,16(1): 110-126.
[11]	CHEN Q M , YANG K X , JIANG H ,et al. Joint beamforming coordination and user selection for CoMP enabled NR-U networks[J]. IEEE Internet of Things Journal, 2021,PP(99): 1.
[12]	KIM Y , JEONG J , AHN S ,et al. Energy and delay guaranteed joint beam and user scheduling policy in 5G CoMP networks[J]. IEEE Transactions on Wireless Communications, 2022,21(4): 2742-2756.
[13]	HE S W , AN Z Y , ZHU J Y ,et al. Cross-layer optimization:joint user scheduling and beamforming design with QoS support in joint transmission networks[J]. arXiv Preprint,arXiv:2203.00934, 2022.
[14]	HE S W , XIONG S W , OU Y Y ,et al. An overview on the application of graph neural networks in wireless networks[J]. IEEE Open Journal of the Communications Society, 2021,2: 2547-2565.
[15]	伏玉笋, 杨根科 . 人工智能在移动通信中的应用:挑战与实践[J]. 通信学报, 2020,41(9): 190-201.
	FU Y S , YANG G K . Application of artificial intelligence in mobile communication:challenge and practice[J]. Journal on Communications, 2020,41(9): 190-201.
[16]	SHEN Y F , SHI Y M , ZHANG J ,et al. A graph neural network approach for scalable wireless power control[C]// Proceedings of 2019 IEEE Globecom Workshops. Piscataway:IEEE Press, 2019: 1-6.
[17]	LEE M Y , YU G D , LI G Y . Graph embedding-based wireless link scheduling with few training samples[J]. IEEE Transactions on Wireless Communications, 2021,20(4): 2282-2294.
[18]	CHEN T R , ZHANG X R , YOU M L ,et al. A GNN-based supervised learning framework for resource allocation in wireless IoT networks[J]. IEEE Internet of Things Journal, 2022,9(3): 1712-1724.
[19]	SHEN Y F , SHI Y M , ZHANG J ,et al. Graph neural networks for scalable radio resource management:architecture design and theoretical analysis[J]. IEEE Journal on Selected Areas in Communications, 2021,39(1): 101-115.
[20]	ZHANG X Y , ZHANG Z M , YANG L X . Joint user association and power allocation in heterogeneous ultra dense network via semi-supervised representation learning[J]. arXiv Preprint,arXiv:2103.15367, 2021.
[21]	GUO J , YANG C Y . Learning power allocation for multi-cell-multi-user systems with heterogeneous graph neural networks[J]. IEEE Transactions on Wireless Communications, 2022,21(2): 884-897.
[22]	FIORETTO F , MAK T W K , VAN HENTENRYCK P . Predicting AC optimal power flows:combining deep learning and Lagrangian dual methods[C]// Proceedings of the AAAI Conference on Artificial Intelligence. Palo Alto:AAAI Press, 2020: 630-637.
[23]	HE S W , AN Z Y , ZHU J Y ,et al. Beamforming design for multiuser uRLLC with finite blocklength transmission[J]. IEEE Transactions on Wireless Communications, 2021,20(12): 8096-8109.
[24]	GILMER J , SCHOENHOLZ S S , RILEY P F ,et al. Neural message passing for quantum chemistry[C]// Proceedings of International Conference on Machine Learning. New York:PMLR, 2017: 1263-1272.
[25]	ZHOU K X , HUANG X , LI Y N ,et al. Towards deeper graph neural networks with differentiable group normalization[J]. Advances in Neural Information Processing Systems, 2020,33: 4917-4928.
[26]	KINGMA D P , BA J . Adam:a method for stochastic optimization[J]. arXiv Preprint,arXiv:1412.6980, 2014.

系统默认参数	参数值
基站总数M	3
基站天线数N_t	8
系统用户数K	20
基站信噪比SNR/dB	10
用户最小速率需求γmin/(bit·(s·Hz)^-1)	2
优化算法的收敛精度	5×10^-3

K	BFS-USBF		ZFBF-US		SCA-USBF		M-JEEPON
K	R₁	R₂	R₁	R₂	R₁	R₂	R₁	R₂
4	12.362	3.162	10.358	2.330	12.096	3.614	11.257	3.032
6	14.912	3.790	12.536	2.852	14.054	4.040	13.565	3.778
8	16.299	4.202	13.764	3.168	15.219	4.392	14.952	4.046
10	17.950	4.497	15.412	3.602	16.567	4.672	16.510	4.140

K	ZFBF-US		SCA-USBF		M-JEEPON
K	R₁	R₂	R₁	R₂	R₁	R₂
5	31.756	4.664	33.098	4.924	32.196	4.980
10	50.821	8.754	53.245	9.886	53.221	9.608
20	67.523	15.326	74.392	17.224	77.148	14.764
30	68.598	19.928	81.598	20.129	89.140	16.724

r_min	ZFBF-US		SCA-USBF		M-JEEPON
r_min	R₁	R₂	R₁	R₂	R₁	R₂
1	64.607	15.458	66.953	19.992	73.248	14.718
2	67.523	15.326	74.392	17.224	77.148	14.764
3	71.765	15.174	76.033	14.85	80.955	14.540
4	77.013	14.998	80.755	14.342	84.051	14.314
5	82.885	14.304	84.124	13.272	88.101	13.112

基于图神经网络的联合用户调度与波束成形优化算法

GNN-based optimization algorithm for joint user scheduling and beamforming

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献 26

相关文章 15

Metrics

推荐阅读 0

[1]	杨龙, 赵丽, 周雨晨, 贺冰涛, 陈健. 缓存辅助的协作NOMA携能传输[J]. 通信学报, 2023, 44(6): 77-89.
[2]	陈晋音, 熊海洋, 马浩男, 郑雅羽. 基于对比学习的图神经网络后门攻击防御方法[J]. 通信学报, 2023, 44(4): 154-166.
[3]	张雷, 王玉, 田建杰, 张琳, 章天骄. 基于IRS辅助的MIMO车联网系统联合波束成形设计[J]. 通信学报, 2023, 44(2): 59-69.
[4]	曾嵘, 杭潇. 车联网环境下可重构智能反射面辅助无线信道估计算法[J]. 通信学报, 2022, 43(8): 142-150.
[5]	郭海燕, 杨震, 邹玉龙, 吕斌, 冯蕴天, 赵玉娟. 基于主被动波束成形联合优化的双RIS辅助抗干扰通信方法[J]. 通信学报, 2022, 43(7): 21-30.
[6]	冷涛, 蔡利君, 于爱民, 朱子元, 马建刚, 李超飞, 牛瑞丞, 孟丹. 基于系统溯源图的威胁发现与取证分析综述[J]. 通信学报, 2022, 43(7): 172-188.
[7]	王志勤, 江甲沫, 刘沛西, 曹晓雯, 李阳, 韩凯峰, 杜滢, 朱光旭. 6G联邦边缘学习新范式：基于任务导向的资源管理策略[J]. 通信学报, 2022, 43(6): 16-27.
[8]	李中捷, 熊吉源, 高伟, 韦金迎. 分布式IRS辅助毫米波MU-MISO系统联合波束成形设计[J]. 通信学报, 2022, 43(4): 216-226.
[9]	吴翼腾, 刘伟, 于洪涛. 图神经网络的标签翻转对抗攻击[J]. 通信学报, 2021, 42(9): 65-74.
[10]	张更新, 王运峰, 丁晓进, 洪涛, 刘子威, 张晨. 卫星互联网若干关键技术研究[J]. 通信学报, 2021, 42(8): 1-14.
[11]	王哲, 李陶深, 葛丽娜, 张桂芬, 吴敏. 基于深度学习的传感云sink节点最优能效SWIPT波束成形设计[J]. 通信学报, 2021, 42(7): 176-188.
[12]	陶梅霞, 王栋, 孙瑞, 张乃夫. 联邦学习中基于时分多址接入的用户调度策略[J]. 通信学报, 2021, 42(6): 1-29.
[13]	朱政宇, 徐金雷, 孙钢灿, 王宁, 郝万明. 基于IRS辅助的SWIPT物联网系统安全波束成形设计[J]. 通信学报, 2021, 42(4): 185-193.
[14]	刘晨曦, 王东, 陈慧玲, 李仁发. 多源异构数据融合的城市私家车流量预测研究[J]. 通信学报, 2021, 42(3): 54-64.
[15]	林钰达,金梁,周游,楼洋明. 噪声不确定时基于波束成形的隐蔽无线通信性能分析[J]. 通信学报, 2020, 41(7): 49-58.