面向不确定CSI随机接入网络的深度稳健资源分配

doi:10.11959/j.issn.1000-436x.2020148

通信学报 ›› 2020, Vol. 41 ›› Issue (7): 29-37.doi: 10.11959/j.issn.1000-436x.2020148

面向不确定CSI随机接入网络的深度稳健资源分配

吴伟华¹,柴冠华¹,杨清海¹,刘润滋²

¹ 西安电子科技大学通信工程学院，陕西西安 710071
² 西安建筑科技大学信息与控制工程学院，陕西西安 710055

修回日期:2020-06-26 出版日期:2020-07-25 发布日期:2020-08-01
作者简介:吴伟华（1988- ），男，河北石家庄人，博士，西安电子科技大学讲师，主要研究方向为无线资源分配、人工智能、随机网络优化及其在LTE-U网络中的应用|柴冠华（1996- ），男，山西大同人，西安电子科技大学博士生，主要研究方向为人工智能、网络资源分配|杨清海（1976- ），男，山东高密人，博士，西安电子科技大学教授，主要研究方向为自主通信、内容交付网络和LTE-A技术等|刘润滋（1988- ），女，山东潍坊人，博士，西安建筑科技大学副教授，主要研究方向为无线网络、空间网络中的资源管理和性能分析
基金资助:
国家自然科学基金资助项目(61801365);国家自然科学基金资助项目(61701365);国家自然科学基金资助项目(61971327);中国博士后科学基金资助项目(2018M643581);陕西省自然科学基金资助项目(2019JQ-152);陕西省自然科学基金资助项目(2020JQ-686);中央高校基本科研业务费专项资金资助项目;陕西省博士后科研基金资助项目

Deep and robust resource allocation for random access network based with imperfect CSI

Weihua WU¹,Guanhua CHAI¹,Qinghai YANG¹,Runzi LIU²

¹ School of Telecommunications Engineering,Xidian University,Xi’an 710071,China
² School of Information and Control Engineering,Xi’an University of Architecture and Technology,Xi’an 710055,China

Revised:2020-06-26 Online:2020-07-25 Published:2020-08-01
Supported by:
The National Natural Science Foundation of China(61801365);The National Natural Science Foundation of China(61701365);The National Natural Science Foundation of China(61971327);China Post-doctoral Science Foundation(2018M643581);The Natural Science Foundation of Shaanxi Province(2019JQ-152);The Natural Science Foundation of Shaanxi Province(2020JQ-686);Research Funds for the Central Universities;Postdoctoral Foundation of Shaanxi Province

摘要/Abstract

摘要：

针对无线随机接入网络中通信信道状态信息（C-CSI）和干扰信道状态信息（I-CSI）均不确定的情况，提出了一种深度稳健资源分配架构。该资源分配架构将无线网络的资源优化目标看作一个学习问题，利用深度神经网络（DNN）以无监督的方式学习最优资源分配策略。通过将不确定的CSI建模为椭圆形状的不确定性集合，提出了一个由2个DNN级联构成的网络结构，第一个是不确定的CSI处理单元，第二个是功率控制单元。然后，提出了一种交替迭代训练算法用于联合训练2个级联的DNN单元。最后，仿真比较了稳健学习策略和非稳健学习策略下的网络性能，验证了所提算法的有效性。

关键词: 深度神经网络, 随机接入网络, 稳健优化, 无线资源分配

Abstract:

A deep and robust resource allocation framework was proposed for the random access based wireless networks,where both the communication channel state information (C-CSI) and the interference channel state information (I-CSI) were uncertain.The proposed resource allocation framework considered the optimization objective of wireless networks as a learning problem and employs deep neural network (DNN) to approximate optimal resource allocation policy through unsupervised manner.By modeling the uncertainties of CSI as ellipsoid sets,two concatenated DNN units were proposed,where the first was uncertain CSI processing unit and the second was the power control unit.Then,an alternating iterative training algorithm was developed to jointly train the two concatenated DNN units.Finally,the simulations verify the effectiveness of the proposed robust leaning approach over the nonrobust one.

Key words: deep neural network, random access network, robust optimization, wireless resource allocation

中图分类号:

TP393

吴伟华,柴冠华,杨清海,刘润滋. 面向不确定CSI随机接入网络的深度稳健资源分配[J]. 通信学报, 2020, 41(7): 29-37.

Weihua WU,Guanhua CHAI,Qinghai YANG,Runzi LIU. Deep and robust resource allocation for random access network based with imperfect CSI[J]. Journal on Communications, 2020, 41(7): 29-37.

图/表 8

图1

图2

图3

表1

图4

图5

图6

图7

参考文献 24

[1]	HUANG Y , CHEN Y , HOU Y T ,et al. Recent advances of LTE/Wi-Fi coexistence in unlicensed spectrum[J]. IEEE Network, 2018,32(2): 107-113.
[2]	ZHENG K , ZHENG Q , CHATZIMISIOS P ,et al. Heterogeneous vehicular networking:a survey on architecture,challenges,and solutions[J]. IEEE Communications Surveys ＆ Tutorials, 2015,17(4): 2377-2396.
[3]	钱志鸿, 王雪 . 面向5G通信网的D2D技术综述[J]. 通信学报, 2016,37(7): 1-14.
	QIAN Z H , WANG X . Reviews of D2D technology for 5G communication networks[J]. Journal on Communications, 2016,37(7): 1-14.
[4]	WU W , YANG Q , LIU R ,et al. Protocol design and resource allocation for LTE-U system utilizing licensed and unlicensed bands[J]. IEEE Access, 2019,7: 67068-67080.
[5]	WU W , YANG Q , LIU R ,et al. Online spectrum partitioning for LTE-U and WLAN coexistence in unlicensed spectrum[J]. IEEE Transactions on Communication, 2020,68(1): 506-520.
[6]	YU W , GINIS G , CIOFFIFI J M . Distributed multiuser power control for digital subscriber lines[J]. IEEE Journal on Selected Areas in Communications, 2002,20(5): 1105-1115.
[7]	SHI Q , RAZAVIYAYN M , LUO Z Q ,et al. An iteratively weighted MMSE approach to distributed sum-utility maximization for a MIMO interfering broadcast channel[J]. IEEE Transactions on Signal Processing, 2011,59(9): 4331-4340.
[8]	SUN H , CHEN X,SHIQ ,et al. Learning to optimize:training deep neural networks for wireless resource management[C]// 2017 IEEE 18th International Workshop on Signal Processing Advances in Wireless Communications. Piscataway:IEEE Press, 2017: 1-6.
[9]	HEE W , CHO D H , KIM M . Resource allocation for multi-channel underlay cognitive radio network based on deep neural network[J]. IEEE Communication Letter, 2018,22(1): 1942-1945.
[10]	DMER S , HOYDIS C J , BRINK S T . Deep learning based communication over the air[J]. IEEE Journal of Selected Topics in Signal Processing, 2018,12(1): 132-143.
[11]	EISEN M , ZHANG C L , CHAMMON F O ,et al. Learning optimal resource allocations in wireless systems[J]. IEEE Transactions on Signal Processing, 2019,67(10): 2775-2790.
[12]	LEE H , LEE S H , QUEK T Q . Deep learning for distributed optimization:applications to wireless resource management[J]. IEEE Journal on Selected Areas in Communications, 2019,(PP):1.
[13]	廖晓闽, 严少虎, 石嘉 ,等. 基于深度强化学习的蜂窝网资源分配算法[J]. 通信学报, 2019,40(2): 11-18.
	LIAO X M , YAN S H , SHI J ,et al. Deep reinforcement learning based resource allocation algorithm in cellular networks[J]. Journal on Communications, 2019,40(2): 11-18.
[14]	LEE W , KIM M , CHO D . Transmit power control using deep neural network for underlay device-to-device communication[J]. IEEE Communications Letters, 2019,8(1): 141-144.
[15]	KERRET P , GESBERT D , RILIPPONE M . Team deep neural networks for interference channels[C]// 2018 IEEE International Conference on Communications Workshops. Piscataway:IEEE Press, 2018: 1-6.
[16]	AI-KHASIB T , SHENOUDA M B , LAMPE L . Dynamic spectrum management for multiple-antenna cognitive radio systems:Designs with imperfect CSI[J]. IEEE Transactions on Wireless Communications, 2011,10(9): 2850-2859.
[17]	MPLARI N , PARSAEEFARD S , AZMI P ,et al. Robust ergodic uplink resource allocation in underlay OFDMA cognitive radio networks[J]. IEEE Transactions on Mobile Computing, 2016,15(2): 419-431.
[18]	PARSAEEFATD S , SHARAFAT A R . Robust worst-case interference control in underlay cognitive radio networks[J]. IEEE Transactions on Vehicular Technology, 2012,61(8): 3731-3745.
[19]	SAGARI S , BAYSTING S , SAHA D ,et al. Coordinated dynamic spectrum management of LTE-U and Wi-Fi networks[C]// 2015 IEEE International Symposium on Dynamic Spectrum Access Networks. Piscataway:IEEE Press, 2015: 1-12.
[20]	LIEW S C , KAI C H , LEUNG H C ,et al. Back-of-the-envelope computation of throughput distributions in CSMA wireless networks[J]. IEEE Transactions on Mobile Computing, 2010,9(9): 1319-1331.
[21]	MALLICK S , DEVARAJAN R , LOODDRICHEH R A ,et al. Robust resource optimization for cooperative cognitive radio networks with imperfect CSI[J]. IEEE Transactions on Wireless Communications, 2014,14(2): 907-920.
[22]	SUN H , CHEN X , SHI Q ,et al. Learning to optimize:Training deep neural networks for interference management[J]. IEEE Transactions Signal Process, 2018,66(20): 5438-5453.
[23]	RIBEIRO A . Optimal resource allocation in wireless communication and networking[J]. EURASIP Journal on Wireless Communications and Networking, 2012,2012(1): 272-285.
[24]	YIN R , YU G , MAAREF A ,et al. A framework for co-channel interference and collision probability tradeoff in LTE licensed-assisted access networks[J]. IEEE Transactions on Wireless Communications, 2016,15(9): 6078-6090.

网络参数	数值大小
CSMA链路个数	6
设备之间的距离d	10～100 m随机取值
带宽B/MHz	6
传输效率a	0.6
路径损耗PL	-15.3-50lg(d[m])
高斯白噪声功率谱密度N₀/(dBm·Hz^-1)	-174
干扰阈值Q/dBm	33
最大发射功率P_max/W	10

面向不确定CSI随机接入网络的深度稳健资源分配

Deep and robust resource allocation for random access network based with imperfect CSI

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 24

相关文章 11

Metrics

推荐阅读 0

[1]	林云, 徐怀韬, 王森, 张思成, 庄龙. 基于特征融合的通信语音干扰效果客观评估[J]. 通信学报, 2023, 44(3): 105-116.
[2]	王莉, 魏青, 徐连明, 沈渊, 张平, 费爱国. 面向通信-导航-感知一体化的应急无人机网络低能耗部署研究[J]. 通信学报, 2022, 43(7): 1-20.
[3]	熊金波, 周永洁, 毕仁万, 万良, 田有亮. 边缘协同的轻量级隐私保护分类框架[J]. 通信学报, 2022, 43(1): 127-137.
[4]	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.
[5]	郝万明, 谢金坤, 孙钢灿, 朱政宇, 周一青. 基于无线供能反向散射通信系统安全性的稳健机会约束优化算法设计[J]. 通信学报, 2021, 42(3): 100-110.
[6]	傅友华, 陈栋. 混合智能反射表面结构辅助的毫米波通信信道估计[J]. 通信学报, 2021, 42(10): 189-196.
[7]	张思成,林云,涂涯,ShiwenMao. 基于轻量级深度神经网络的电磁信号调制识别技术[J]. 通信学报, 2020, 41(11): 12-21.
[8]	胡铮,袁浩,朱新宁,倪万里. 面向5G需求的人群流量预测模型研究[J]. 通信学报, 2019, 40(2): 1-10.
[9]	桂冠,王禹,黄浩. 基于深度学习的物理层无线通信技术：机遇与挑战[J]. 通信学报, 2019, 40(2): 19-23.
[10]	姜永,陈山枝,胡博. 异构无线网络中基于Stackelberg博弈的分布式定价和资源分配算法[J]. 通信学报, 2013, 34(1): 61-68.
[11]	刘彩霞,邬江兴,俞定玖,程东年. Cdma2000-1x系统分组数据去激活定时器的优化设计方法[J]. 通信学报, 2005, 26(2): 92-99.