基于深度图嵌入的无人机自组网链路预测

doi:10.11959/j.issn.1000-436x.2021083

通信学报 ›› 2021, Vol. 42 ›› Issue (7): 137-149.doi: 10.11959/j.issn.1000-436x.2021083

基于深度图嵌入的无人机自组网链路预测

舒坚¹, 王启宁¹, 刘琳岚²

¹ 南昌航空大学软件学院，江西南昌 330063
² 南昌航空大学信息工程学院，江西南昌 330063

修回日期:2020-12-01 出版日期:2021-07-25 发布日期:2021-07-01
作者简介:舒坚（1964− ），男，江西南昌人，南昌航空大学教授、硕士生导师，主要研究方向为分布式系统、软件工程等
王启宁（1996− ），男，河北石家庄人，南昌航空大学硕士生，主要研究方向为分布式系统
刘琳岚（1968− ），女，湖南东安人，南昌航空大学教授、硕士生导师，主要研究方向为无线传感器网络、分布式系统等
基金资助:
国家自然科学基金资助项目(62062050);国家自然科学基金资助项目(61762065);国家自然科学基金资助项目(61962037);江西省自然科学基金资助项目(20202BABL202039);江西省研究生创新专项资金资助项目(YC2019-S355)

UAV ad hoc network link prediction based on deep graph embedding

Jian SHU¹, Qining WANG¹, Linlan LIU²

¹ School of Software, Nanchang Hangkong University, Nanchang 330063, China
² School of Information Engineering, Nanchang Hangkong University, Nanchang 330063, China

Revised:2020-12-01 Online:2021-07-25 Published:2021-07-01
Supported by:
The National Natural Science Foundation of China(62062050);The National Natural Science Foundation of China(61762065);The National Natural Science Foundation of China(61962037);The Natural Science Foundation of Jiangxi Province(20202BABL202039);The Innovation Foundation for Postgraduate Student of Jiangxi Province(YC2019-S355)

摘要/Abstract

摘要：

针对无人机自组网的拓扑时变、节点移动、间歇性连接等特点，提出用时序化图嵌入模型对预处理后的无人机自组网进行表征，基于线性概率计算采样间隔以提高采样效率，将网络结构特征映射为节点间关系，并采用对抗训练提取节点上下文语义特征。利用长短期记忆网络提取无人机自组网的时序特征，预测下一时刻的网络连接情况。采用AUC、MAP、Error Rate作为评价指标。Ns-3仿真实验表明，与Node2vec、DDNE、E-LSTM-D等方法相比，所提方法具有更高的预测准确率。

关键词: 无人机自组网, 图嵌入, 链路预测, 长短期记忆网络

Abstract:

Aiming at the characteristics of the UAV ad hoc network (UAANET), such as topological temporal-varying, node mobility and intermittent connection, a temporal graph embedding model was proposed to present the preprocessed UAANET.To improve the sampling efficiency, the sampling interval was calculated based on linear probability.The network structure features were mapped to the relationship between nodes, and the contextual semantic features of nodes were extracted by adversarial training.With the help of long and short-term memory network, the temporal characteristics of the UAANET were extracted to predict the connection at the next moment.AUC, MAP, and Error Rate were employed as evaluation indexes.The simulation experiments based on NS-3 show that compared with Node2vec, DDNE and E-LSTM-D, the proposed method has a better accuracy.

Key words: UAV ad hoc network, graph embedding, link prediction, long short-term memory network

中图分类号:

TP393

舒坚, 王启宁, 刘琳岚. 基于深度图嵌入的无人机自组网链路预测[J]. 通信学报, 2021, 42(7): 137-149.

Jian SHU, Qining WANG, Linlan LIU. UAV ad hoc network link prediction based on deep graph embedding[J]. Journal on Communications, 2021, 42(7): 137-149.

图/表 23

图1

图2

表1

符号及描述"

符号	描述
$G = {G_{t}}_{1}^{T}$	动态无人机自组网
$G_{t} = {V_{t}, E_{t}}$	t时刻的网络快照
V ,E	点集和边集
$Γ (v_{i})$	节点v_i在静态快照内的邻居列表
$γ (v_{i})$	节点v_i在游走路径上的邻居列表
$S_{t} ， {\hat{S}}_{t}$	S _t为输入的一组快照， ${\hat{S}}_{t}$ 为嵌入矩阵
N ,n	网络快照的个数N 和节点总数n
$e_{i j}, w_{i j}$	节点v_i和v_j 间的连接和权重
$u_{i}$	节点v_i的嵌入向量
$a_{i j}, {\hat{a}}_{i j}$	连接的真实值a_ij 和预测值 ${\hat{a}}_{i j}$
$A_{t}, {\hat{A}}_{t}$	网络的真实值 $A_{t}$ 和预测值 ${\hat{A}}_{t}$
$W_{f}, W_{i}, W_{c}, W_{o}$	LSTM单元中的权重矩阵
$b_{f}, b_{i}, b_{c}, b_{o}$	LSTM单元中的偏置
$W_{d}, b_{d}$	解码器中的权重矩阵和偏置
d (i,j)	节点v_i和v_j 间的距离
D_i	节点v_i的最大传输半径
β ,τ,ρ,ε	模型超参数

表1

图3

图4

图5

图6

图7

图8

表2

图9

图10

图11

图12

图13

图14

图15

图16

图17

图18

表3

链路预测效果对比"

模型	AUC	MAP	Error Rate
CN	0.879 367	0.104 817	3.997 406
Katz	0.893 730	0.119 869	3.756 434
Node2vec-LRCV	0.921 239	0.151 614	3.017 641
Node2vec-LSTM	0.912 372	0.398 330	3.452 423
DDNE	0.959 129	0.654 284	2.375 884
E-LSTM-D	0.977 672	0.745 849	1.384 087
CGEM-LSTM	$0 . 986802$	$0 . 794936$	$0 . 982520$

表3

表4

嵌入效果对比（逻辑回归分类器）"

模型	AUC	MAP	Error Rate
DeepWalk-LRCV	0.869 422	0.269 422	4.137 679
LINE-LRCV	0.798 097	0.198 097	$2 . 025004$
Node2vec-LRCV	$0 . 921239$	0.151 614	3.017 641
CGEM-LRCV	0.916 562	$0 . 296562$	2.645 962

表4

表5

嵌入效果对比（LSTM）"

模型	AUC	MAP	Error Rate
DeepWalk-LSTM	0.897 139	0.294 358	1.423 943
LINE-LSTM	0.857 157	0.275 965	2.361 550
Node2vec-LSTM	0.912 372	0.398 330	3.452 423
CGEM-LSTM	$0 . 986802$	$0 . 794936$	$0 . 982520$

表5

参考文献 23

[1]	ZENG Y , ZHANG R , LIM T J . Wireless communications with unmanned aerial vehicles:opportunities and challenges[J]. IEEE Communications Magazine, 2016,54(5): 36-42.
[2]	任佳智, 田辉, 范绍帅 ,等. 基于用户偏好预测的无人机部署和缓存策略[J]. 通信学报, 2020,41(6): 1-13.
	REN J Z , TIAN H , FAN S S ,et al. UAV deployment and caching scheme based on user preference prediction[J]. Journal on Communications, 2020,41(6): 1-13.
[3]	熊永平, 孙利民, 牛建伟 ,等. 机会网络[J]. 软件学报, 2009,20(1): 124-137.
	XIONG Y P , SUN L M , NIU J W ,et al. Opportunistic networks[J]. Journal of Software, 2009,20(1): 124-137.
[4]	HE Y X , TANG X , ZHANG R N ,et al. A course-aware opportunistic routing protocol for FANETs[J]. IEEE Access, 2019,7: 144303-144312.
[5]	NAGADIVYA S , MANOHARAN R . Energy efficient Markov prediction based opportunistic routing (Eempor) for wireless sensor networks[C]// 2019 IEEE International Conference on System,Computation,Automation and Networking. Piscataway:IEEE Press, 2019: 1-6.
[6]	吕琳媛 . 复杂网络链路预测[J]. 电子科技大学学报, 2010,39(5): 651-661.
	LYU L Y . Link prediction on complex networks[J]. Journal of University of Electronic Science and Technology of China, 2010,39(5): 651-661.
[7]	ISMAIL G , SULE G , ZEHRA C . Link prediction using time series of neighborhood-based node similarity scores[J]. Data Mining and Knowledge Discovery, 2016,30(1): 147-180.
[8]	CHUAN P M , SON L H , ALI M ,et al. Link prediction in co-authorship networks based on hybrid content similarity metric[J]. Applied Intelligence, 2018,48(8): 2470-2486.
[9]	RAFIEE S , SALAVATI C , ABDOLLAHPOURI A . CNDP:link prediction based on common neighbors degree penalization[J]. Physica A:Statistical Mechanics and Its Applications, 2020,539:122950.
[10]	PAN L , ZHOU T , LYU L ,et al. Predicting missing links and identifying spurious links via likelihood analysis[J]. Scientific Reports, 2016,6(1): 1-10.
[11]	PECH R , HAO D , LEE Y L ,et al. Link prediction via linear optimization[J]. Physica A:Statistical Mechanics and Its Applications, 2019,528(8): 121319-121348.
[12]	WANG P , XU B W , WU Y R ,et al. Link prediction in social networks:the state-of-the-art[J]. Science China Information Sciences, 2015,58(1): 1-38.
[13]	WANG D X , CUI P , ZHU W W . Structural deep network embedding[C]// Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York:ACM Press, 2016: 1225-1234.
[14]	PAN S R , HU R Q , LONG G D ,et al. Adversarially regularized graph autoencoder for graph embedding[C]// Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence. Saarland:DBLP, 2018: 2609-2615.
[15]	LI T S , ZHANG J W , YU P S ,et al. Deep dynamic network embedding for link prediction[J]. IEEE Access, 2018,6: 29219-29230.
[16]	CHEN J Y , ZHANG J , XU X H ,et al. E-LSTM-D:a deep learning framework for dynamic network link prediction[J]. IEEE Transactions on Systems,Man,and Cybernetics:Systems, 2021,51(6): 3699-3712.
[17]	PEROZZI B , AL-RFOU R ,, SKIENA S . DeepWalk:online learning of social representations[C]// Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York:ACM Press, 2014: 701-710.
[18]	TANG J , QU M , WANG M Z ,et al. LINE:large-scale information network embedding[C]// Proceedings of the 24th International Conference on World Wide Web. New York:ACM Press, 2015: 1067-1077.
[19]	GROVER A , LESKOVEC J . Node2vec:scalable feature learning for networks[C]// Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York:ACM Press, 2016: 855-864.
[20]	DAI Q , SHEN X , ZHANG L ,et al. Adversarial training methods for network embedding[C]// The World Wide Web Conference 2019. New York:ACM Press, 2019: 329-339.
[21]	DU L , WANG Y , SONG G J ,et al. Dynamic network embedding:an extended approach for skip-gram based network embedding[C]// Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence. Saarland:DBLP, 2018: 2086-2092.
[22]	KIPF T , FETAYA E , WANG K C ,et al. Neural relational inference for interacting systems[C]// International Conference on Machine Learning. Stockholm:PMLR, 2018: 2688-2697.
[23]	BOUACHIR O , ABRASSART A , GARCIA F ,et al. A mobility model for UAV ad hoc network[C]// 2014 International Conference on Unmanned Aircraft Systems. Piscataway:IEEE Press, 2014: 383-388.

无人机数量/架	飞行速度/(km.h^-2)	侦察领域/km²	盘旋半径/m	时长/h
100	126	25 000	15	2

基于深度图嵌入的无人机自组网链路预测

UAV ad hoc network link prediction based on deep graph embedding

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 23

参考文献 23

相关文章 12

Metrics

推荐阅读 0

[1]	马玲, 樊漆亮, 许婷, 郭冠琛, 张圣林, 孙永谦, 张玉志. 基于强化学习的在线离线混部云环境下的调度框架[J]. 通信学报, 2023, 44(6): 90-102.
[2]	孙雁飞, 尹嘉峥, 亓晋, 胡筱旋, 陈梦婷, 董振江. 基于动态图嵌入的车联网拓扑控制[J]. 通信学报, 2022, 43(6): 133-142.
[3]	肖振宇, 刘珂, 朱立鹏. 无人机机间毫米波阵列通信技术[J]. 通信学报, 2022, 43(10): 196-209.
[4]	顾秋阳, 吴宝, 池仁勇. 基于高阶路径相似度的复杂网络链路预测方法[J]. 通信学报, 2021, 42(7): 61-69.
[5]	黄瑞章, 靳文繁, 陈艳平, 秦永彬, 郑庆华. 基于Highway-BiLSTM网络的汉语谓语中心词识别研究[J]. 通信学报, 2021, 42(1): 100-107.
[6]	吴翼腾,于洪涛,黄瑞阳,李华巍. 采用组合方法进行链路预测的理论极限研究[J]. 通信学报, 2020, 41(6): 34-50.
[7]	刘树新,李星,陈鸿昶,王凯. 基于资源传输匹配度的复杂网络链路预测方法[J]. 通信学报, 2020, 41(6): 70-79.
[8]	丁恩杰,刘忠育,刘亚峰,郁万里. 基于多维度和多模态信息的视频描述方法[J]. 通信学报, 2020, 41(2): 36-43.
[9]	顾秋阳, 琚春华, 吴功兴. 基于子图演化与改进蚁群优化算法的社交网络链路预测方法[J]. 通信学报, 2020, 41(12): 21-35.
[10]	王丽娜,郭晓东,汪润. 面向中文用户评论的自动化众包攻击方法[J]. 通信学报, 2019, 40(6): 1-13.
[11]	吴祖峰，梁棋，刘峤，秦志光. 基于AdaBoost的链路预测优化算法[J]. 通信学报, 2014, 35(3): 13-123.
[12]	吴祖峰,梁棋,刘峤,秦志光. 基于AdaBoost的链路预测优化算法[J]. 通信学报, 2014, 35(3): 116-123.