Distributed power allocation algorithm based on graph convolutional network for D2D communication systems

doi:10.11772/j.issn.1001-9081.2023081221

Journal of Computer Applications ›› 2024, Vol. 44 ›› Issue (9): 2855-2862.DOI: 10.11772/j.issn.1001-9081.2023081221

• Network and communications • Previous Articles Next Articles

Distributed power allocation algorithm based on graph convolutional network for D2D communication systems

Chuanlin PANG¹, Rui TANG²(), Ruizhi ZHANG³, Chuan LIU², Jia LIU², Shibo YUE¹

^1.College of Computer Science and Cyber Security，Chengdu University of Technology，Chengdu Sichuan 610059，China
^2.School of Electronic Information Engineering，China West Normal University，Nanchong Sichuan 637002，China
^3.School of Information and Communication Engineering，University of Electronic Science and Technology of China，Chengdu Sichuan 611731，China

Received:2023-09-08 Revised:2023-10-23 Accepted:2023-11-02 Online:2024-09-14 Published:2024-09-10
Contact: Rui TANG
About author:PANG Chuanlin， born in 1997， M. S. candidate. His research interests include deep reinforcement learning algorithms.
ZHANG Ruizhi， born in 1999， M. S. candidate. His research interests include generalized optimization algorithms.
LIU Chuan， born in 1991， M. S.， lecturer. His research interests include PHY layer design in radio communication.
LIU Jia， born in 1995， M. S.， lecturer. His research interests include flight control design， image processing.
YUE Shibo， born in 2000， M. S. candidate. His research interests include resource allocation in unmanned aerial vehicular communication.
Supported by:
This work is partially supported by National Natural Science Foundation of China （62301450）； Sichuan Provincial Natural Science Foundation （2024NSFSC1420）； Regional Innovation Cooperation Project of Science and Technology Department of Sichuan Province （2022YFQ0017）； Fundamental Research Funds of China West Normal University （22kE007，473762）.

D2D通信系统中基于图卷积网络的分布式功率控制算法

庞川林¹, 唐睿²(), 张睿智³, 刘川², 刘佳², 岳士博¹

^1.成都理工大学计算机与网络安全学院, 成都 610059
^2.西华师范大学电子信息工程学院, 四川南充 637002
^3.电子科技大学信息与通信工程学院, 成都 611731

通讯作者: 唐睿
作者简介:庞川林（1997—），男，四川南充人，硕士研究生，主要研究方向：深度强化学习算法；
张睿智（1999—），男，山东济宁人，硕士研究生，主要研究方向：泛化优化算法；
刘川（1991—），男，四川南充人，讲师，硕士，主要研究方向：无线通信中PHY层设计；
刘佳（1995—），男，四川南充人，讲师，硕士，主要研究方向：飞控设计、图像处理；
岳士博（2000—），男，四川巴中人，硕士研究生，CCF会员，主要研究方向：无人机通信中资源分配。
基金资助:
国家自然科学基金资助项目(62301450);四川省自然科学基金资助项目(2024NSFSC1420);四川省科技厅区域创新合作项目(2022YFQ0017);西华师范大学校级基本科研业务费资金资助项目(22kE007)

Abstract

Abstract:

In order to effectively control the co-channel interference in Device-to-Device （D2D） communication system while reducing the implementation complexity of the system， a Graph Convolutional Network （GCN）-based distributed power allocation algorithm was proposed to maximize the weighted sum rate of all D2D links. Firstly， the system topology was built into a graph model， and the characteristics of nodes and edges as well as the message-passing manners were defined. Then， the unsupervised learning model was used to train the model parameters in the GCN. After the offline training， each D2D link was able to obtain the optimal power allocation strategy in a distributed manner based on local channel state information and the interaction with neighboring nodes. Experimental results show that compared with the optimization theory-based algorithm， the proposed algorithm cuts down the running time by 97.41% while suffering only 3.409% weighted sum rate loss； and compared with the deep reinforcement learning theory-based algorithm， the proposed algorithm has better generalization ability and is stable under different setting of parameters.

Key words: Device-to-Device (D2D) communication, interference coordination, Graph Convolutional Network (GCN), power allocation, distributed algorithm

摘要：

为有效控制终端直通（D2D）通信系统中的同频干扰并降低它的实现复杂度，提出一种基于图卷积网络（GCN）的分布式功率控制算法，旨在最大化所有D2D链路的加权和速率。首先将系统拓扑结构建模为图模型并定义节点和边的特征以及消息传递方式，随后借助无监督学习模型训练GCN中的模型参数。离线训练后，每条D2D链路可根据局部信道状态信息以及与相邻节点的交互过程来分布式地得到最佳功率控制策略。实验结果表明，相较于基于优化理论的算法，所提算法缩短了97.41%的运算时间且仅损失了3.409%的加权和速率；相较于基于深度强化学习理论的算法，所提算法具有良好的泛化能力，在不同参数设置下表现更加稳定。

关键词: 终端直通通信, 干扰协调, 图卷积网络, 功率控制, 分布式算法

CLC Number:

TP393

Chuanlin PANG, Rui TANG, Ruizhi ZHANG, Chuan LIU, Jia LIU, Shibo YUE. Distributed power allocation algorithm based on graph convolutional network for D2D communication systems[J]. Journal of Computer Applications, 2024, 44(9): 2855-2862.

庞川林, 唐睿, 张睿智, 刘川, 刘佳, 岳士博. D2D通信系统中基于图卷积网络的分布式功率控制算法[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2855-2862.

Figures/Tables 15

Fig. 1 Structure of D2D communication system

Fig. 2 Graph model of D2D communication system

Fig. 3 Structure of GCN-based decision model

Fig. 4 Structure of nonlinear network

Fig. 5 Process of message passing

Fig. 6 Structure of fully connected feedforward DNN

Fig. 7 Offline training flow of GCN

Tab. 1 System parameters

参数	设置1	设置2
小区半径	300 m	500 m
D2D通信对距离	5~50 m	10~60 m
D2D通信对发射功率	0~200 mW	0~200 mW
路径损耗^［32］	$128.1 + 37.6 l g (d)$	$128.1 + 37.6 l g (d)$
阴影衰落	$l o g n o r m a l (0,7)$	$l o g n o r m a l (0,8)$
小尺度衰落	$R a y l e i g h (1)$	$R a y l e i g h (1.1)$
高斯白噪声	-113 dBm	-113 dBm
求和权重	［0.5，1］	［0.5，1］

Tab. 1 System parameters

参数	设置1	设置2
小区半径	300 m	500 m
D2D通信对距离	5~50 m	10~60 m
D2D通信对发射功率	0~200 mW	0~200 mW
路径损耗^［32］	$128.1 + 37.6 l g (d)$	$128.1 + 37.6 l g (d)$
阴影衰落	$l o g n o r m a l (0,7)$	$l o g n o r m a l (0,8)$
小尺度衰落	$R a y l e i g h (1)$	$R a y l e i g h (1.1)$
高斯白噪声	-113 dBm	-113 dBm
求和权重	［0.5，1］	［0.5，1］

Tab. 2 Hyperparameter setting in GCN-based decision model

参数	值
$L 1 (0) (⋅; θ 1 (0))$ 神经元数	8
$L 2 (0) (⋅; θ 2 (0))$ 神经元数	16
$L 3 (t) (⋅; θ 3 (t))$ 图卷积层层数	2
$L 3 (t) (⋅; θ 3 (t))$ 神经元数	（16，24）
图卷积层激活函数	ReLU
$D N N (⋅; θ D N N)$ 隐藏层层数	2
$D N N (⋅; θ D N N)$ 神经元数	（32，16）
隐藏层激活函数	ReLU
输出层激活函数	Sigmoid
优化器	Adam
学习率	0.001

Tab. 2 Hyperparameter setting in GCN-based decision model

参数	值
$L 1 (0) (⋅; θ 1 (0))$ 神经元数	8
$L 2 (0) (⋅; θ 2 (0))$ 神经元数	16
$L 3 (t) (⋅; θ 3 (t))$ 图卷积层层数	2
$L 3 (t) (⋅; θ 3 (t))$ 神经元数	（16，24）
图卷积层激活函数	ReLU
$D N N (⋅; θ D N N)$ 隐藏层层数	2
$D N N (⋅; θ D N N)$ 神经元数	（32，16）
隐藏层激活函数	ReLU
输出层激活函数	Sigmoid
优化器	Adam
学习率	0.001

Fig. 8 Training and testing convergence processes （M=20）

Tab. 3 Online computational complexity of different algorithms

算法	在线运算复杂度	算法	在线运算复杂度
DDPG	$O t m a x ∑ j = 0 J - 1 Y j Y j + 1$	SAC	$O t m a x ∑ j = 0 J - 1 Y j Y j + 1$
FP	$O (T m a x M 3.5)$	本文算法	$O (C C O N V + C D N N)$

Tab. 3 Online computational complexity of different algorithms

算法	在线运算复杂度	算法	在线运算复杂度
DDPG	$O t m a x ∑ j = 0 J - 1 Y j Y j + 1$	SAC	$O t m a x ∑ j = 0 J - 1 Y j Y j + 1$
FP	$O (T m a x M 3.5)$	本文算法	$O (C C O N V + C D N N)$

Tab. 4 Comparison of running time varying with change in number of D2D pairs of different algorithms

算法	M=10	M=15	M=20	M=25	M=30
DDPG	13.57	13.92	14.25	14.56	14.83
SAC	13.48	13.87	14.23	14.53	14.78
FP	358.40	435.80	517.70	615.30	718.60
MO	6.215 $× 105$	1.327 $× 106$	2.736 $× 106$	5.642 $× 106$	1.187 $× 107$
本文算法	11.86	12.05	12.28	12.46	12.73

Tab. 4 Comparison of running time varying with change in number of D2D pairs of different algorithms

算法	M=10	M=15	M=20	M=25	M=30
DDPG	13.57	13.92	14.25	14.56	14.83
SAC	13.48	13.87	14.23	14.53	14.78
FP	358.40	435.80	517.70	615.30	718.60
MO	6.215 $× 105$	1.327 $× 106$	2.736 $× 106$	5.642 $× 106$	1.187 $× 107$
本文算法	11.86	12.05	12.28	12.46	12.73

Fig. 9 Weighted sum rate varying with change in number of D2D pairs （setting 1）

Fig. 10 Weighted sum rate varying with change in number of D2D pairs （setting 2）

Tab. 5 Average transmit power of D2D pairs （M=20）

算法	系统参数设置1	系统参数设置2
算法	系统参数设置1	实验1	实验2
DDPG	112.5	141.7	141.4
SAC	113.8	142.5	142.9
FP	115.6	146.4	146.6
MO	116.9	148.8	148.8
本文算法	114.7	144.3	144.1

References 35

1	ZHANG S， LIU J， GUO H， et al. Envisioning device-to-device communications in 6G ［J］. IEEE Network， 2020， 34（3）： 86-91.
2	HAN Y， LIU L， DUAN L， et al. Towards reliable UAV swarm communication in D2D-enhanced cellular networks ［J］. IEEE Transactions on Wireless Communications， 2021， 20（3）： 1567-1581.
3	MAO S， CHU X， WU Q， et al. Intelligent reflecting surface enhanced D2D cooperative computing ［J］. IEEE Wireless Communications Letters， 2021， 10（7）： 1419-1423.
4	李余，何希平，唐亮贵. 基于终端直通通信的多用户计算卸载资源优化决策［J］. 计算机应用， 2022， 42（5）： 1538-1546.
	LI Y， HE X P， TANG L G. Multi-user computation offloading and resource optimization policy based on device-to-device communication ［J］. Journal of Computer Applications， 2022， 42（5）： 1538-1546.
5	CHIANG M， HANDE P， LAN T， et al. Power control in wireless cellular networks ［J］. Foundations and Trends in Networking， 2008， 2（4）： 381-533.
6	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.
7	TANG R， ZHAO J， QU H， et al. User-centric joint admission control and resource allocation for 5G D2D extreme mobile broadband： a sequential convex programming approach ［J］. IEEE Communications Letters， 2017， 21（7）： 1641-1644.
8	TANG R， ZHAO J， QU H. Performance tradeoff between energy conservation and user fairness for D2D communication underlaying cellular networks ［J］. Chinese Journal of Electronics， 2017， 26（3）： 600-607.
9	尼俊红，申振涛，杨会峰. 蜂窝网络下基于max-min公平性的D2D功率分配［J］. 计算机应用， 2017， 37（4）： 945-947.
	NI J H， SHEN Z T， YANG H F. D2D power allocation based on max-min fairness underlying cellular systems ［J］. Journal of Computer Applications， 2017， 37（4）： 945-947.
10	SHEN K， YU W. FPLinQ： a cooperative spectrum sharing strategy for device-to-device communications ［C］// Proceedings of the 2017 IEEE International Symposium on Information Theory. Piscataway： IEEE， 2017： 2323-2327.
11	SHEN K， YU W. Fractional programming for communication systems — part Ⅱ： uplink scheduling via matching ［J］. IEEE Transactions on Signal Processing， 2018， 66（10）： 2631-2644.
12	SUN H， CHEN X， SHI Q， et al. Learning to optimize： training deep neural networks for interference management ［J］. IEEE Transactions on Signal Processing， 2018， 66（20）： 5438-5453.
13	NASIR Y S， GUO D. Multi-agent deep reinforcement learning for dynamic power allocation in wireless networks ［J］. IEEE Journal on Selected Areas in Communications， 2019， 37（10）： 2239-2250.
14	唐睿，何祖涵，张睿智，等. D2D-NOMA系统中混合离线-在线资源分配机制［J］. 信息与控制， 2023， 52（5）：574-587.
	TANG R， HE Z H， ZHANG R Z， et al. Hybrid offline-online resource allocation mechanism for D2D-NOMA systems ［J］. Information and Control， 2023， 52（5）：574-587.
15	张先超，赵耀，叶海军，等. 无线网络多用户干扰下智能发射功率控制算法［J］. 通信学报， 2022， 43（2）： 15-21.
	ZHANG X C， ZHAO Y， YE H J， et al. Intelligent transmit power control algorithm for the multi-user interference of wireless network ［J］. Journal on Communications， 2022， 43（2）： 15-21.
16	BAN T W. A self-regulating power-control scheme using reinforcement learning for D2D communication networks ［J］. Sensors， 2022， 22（13）： No.4894.
17	唐睿，庞川林，张睿智，等. D2D通信增强的蜂窝网络中基于DDPG的资源分配［J］. 计算机应用， 2024， 44（5）： 1562-1569.
	TANG R， PANG C L， ZHANG R Z， et al. DDPG-based resource allocation in D2D communication-empowered cellular network ［J］. Journal of Computer Applications， 2024， 44（5）： 1562-1569.
18	JIA K， CHEN D， SUN X. Soft actor-critic based power control algorithm for anti-jamming in D2D communication ［C］// Proceedings of the 2023 IEEE International Conference on Control， Electronics and Computer Technology. Piscataway： IEEE， 2023： 1-5.
19	ZAPPONE A， DI RENZO M， DEBBAH M. Wireless networks design in the era of deep learning： model-based， AI-based， or both？［J］. IEEE Transactions on Communications， 2019， 67（10）：7331-7376.
20	ZHANG R， TANG R， XU Y， et al. Resource allocation for UAV-assisted NOMA systems with dual connectivity ［J］. IEEE Wireless Communications Letters， 2023， 12（2）： 341-345.
21	SHEN K， YU W. Fractional programming for communication systems — part I： power control and beamforming ［J］. IEEE Transactions on Signal Processing， 2018， 66（10）： 2616-2630.
22	SCHMIDHUBER J. Deep learning in neural networks： an overview［J］. Neural Networks， 2015， 61： 85-117.
23	ZHOU J， CUI G， HU S， et al. Graph neural networks： a review of methods and applications ［J］. AI Open， 2020， 1： 57-81.
24	WU Z， PAN S， CHEN F， et al. A comprehensive survey on graph neural networks ［J］. IEEE Transactions on Neural Networks and Learning Systems， 2021， 32（1）： 4-24.
25	LEE M， YU G， DAI H， et al. Graph neural networks meet wireless communications： motivation， applications， and future directions ［J］. IEEE Wireless Communications， 2022， 29（5）： 12-19.
26	BATTAGLIA P W， HAMRICK J B， BAPST V. Relational inductive biases， deep learning， and graph networks［EB/OL］. ［2023-08-11］. .
27	GILMER J， SCHOENHOLZ S S， RILEY P F. Neural message passing for quantum chemistry ［C］// Proceedings of the 34th International Conference on Machine Learning. New York： JMLR.org， 2017：1263-1272.
28	LeCUN Y， BENGIO Y， HINTON G. Deep learning ［J］. Nature， 2015， 521（7553）： 436-444.
29	ZAFAR A， AAMIR M， MOHD NAWI N， et al. A comparison of pooling methods for convolutional neural networks ［J］. Applied Sciences， 2022， 12（17）： 8643.
30	LESHNO M， LIN V Y， PINKUS A， et al. Multilayer feedforward networks with a nonpolynomial activation function can approximate any function ［J］. Neural Networks， 1993， 6（6）： 861-867.
31	KINGMA D P， BA J L. Adam： a method for stochastic optimization［EB/OL］. （2017-01-30）［2023-08-13］. .
32	TANG R， ZHANG R， XU Y， et al. Energy-efficient optimization algorithm in NOMA-based UAV-assisted data collection systems［J］. IEEE Wireless Communications Letters， 2023， 12（1）： 158-162.
33	TANG R， ZHANG R， ZHAO Y， et al. Power allocation for NOMA-based two-way full-duplex relaying systems ［C］// Proceedings of the 2021 IEEE/CIC International Conference on Communications in China. Piscataway： IEEE， 2021： 1077-1082.
34	TUY H. Monotonic optimization： problems and solution approaches ［J］. SIAM Journal on Optimization， 2000， 11（2）： 464-494.
35	HERBERT S， WASSELL I， LOH T H， et al. Characterizing the spectral properties and time variation of the in-vehicle wireless communication channel ［J］. IEEE Transactions on Communications， 2014， 62（7）： 2390-2399.

[1]	Guixiang XUE, Hui WANG, Weifeng ZHOU, Yu LIU, Yan LI. Port traffic flow prediction based on knowledge graph and spatio-temporal diffusion graph convolutional network [J]. Journal of Computer Applications, 2024, 44(9): 2952-2957.
[2]	Huanhuan LI, Tianqiang HUANG, Xuemei DING, Haifeng LUO, Liqing HUANG. Public traffic demand prediction based on multi-scale spatial-temporal graph convolutional network [J]. Journal of Computer Applications, 2024, 44(7): 2065-2072.
[3]	Shibin LI, Jun GONG, Shengjun TANG. Semi-supervised heterophilic graph representation learning model based on Graph Transformer [J]. Journal of Computer Applications, 2024, 44(6): 1816-1823.
[4]	Rui TANG, Chuanlin PANG, Ruizhi ZHANG, Chuan LIU, Shibo YUE. DDPG-based resource allocation in D2D communication-empowered cellular network [J]. Journal of Computer Applications, 2024, 44(5): 1562-1569.
[5]	Longtao GAO, Nana LI. Aspect sentiment triplet extraction based on aspect-aware attention enhancement [J]. Journal of Computer Applications, 2024, 44(4): 1049-1057.
[6]	Xianfeng YANG, Yilei TANG, Ziqiang LI. Aspect-level sentiment analysis model based on alternating‑attention mechanism and graph convolutional network [J]. Journal of Computer Applications, 2024, 44(4): 1058-1064.
[7]	Kaitian WANG, Qing YE, Chunlei CHENG. Classification method for traditional Chinese medicine electronic medical records based on heterogeneous graph representation [J]. Journal of Computer Applications, 2024, 44(2): 411-417.
[8]	Zucheng WU, Xiaojun WU, Tianyang XU. Image-text retrieval model based on intra-modal fine-grained feature relationship extraction [J]. Journal of Computer Applications, 2024, 44(12): 3776-3783.
[9]	Xinrong HU, Jingxue CHEN, Zijian HUANG, Bangchao WANG, Xun YAO, Junping LIU, Qiang ZHU, Jie YANG. Graph convolution network-based masked data augmentation [J]. Journal of Computer Applications, 2024, 44(11): 3335-3344.
[10]	Nengqiang XIANG, Xiaofei ZHU, Zhaoze GAO. Information diffusion prediction model of prototype-aware dual-channel graph convolutional neural network [J]. Journal of Computer Applications, 2024, 44(10): 3260-3266.
[11]	Yanbo LI, Qing HE, Shunyi LU. Aspect sentiment triplet extraction integrating semantic and syntactic information [J]. Journal of Computer Applications, 2024, 44(10): 3275-3280.
[12]	Wanting JI, Wenyi LU, Yuhang MA, Linlin DING, Baoyan SONG, Haolin ZHANG. Machine reading comprehension event detection based on relation-enhanced graph convolutional network [J]. Journal of Computer Applications, 2024, 44(10): 3288-3293.
[13]	Hanxiao SHI, Leichun WANG. Short-term power load forecasting by graph convolutional network combining LSTM and self-attention mechanism [J]. Journal of Computer Applications, 2024, 44(1): 311-317.
[14]	Yi ZHANG, Gangsheng CAI, Zhenmei WANG. Long non-coding RNA-disease association prediction model based on semantic and global dual attention mechanism [J]. Journal of Computer Applications, 2023, 43(7): 2125-2132.
[15]	Yuan WEI, Yan LIN, Shengnan GUO, Youfang LIN, Huaiyu WAN. Prediction of taxi demands between urban regions by fusing origin-destination spatial-temporal correlation [J]. Journal of Computer Applications, 2023, 43(7): 2100-2106.

Distributed power allocation algorithm based on graph convolutional network for D2D communication systems

D2D通信系统中基于图卷积网络的分布式功率控制算法

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 15

References 35

Related Articles 15

Recommended Articles

Metrics