融合Stackelberg博弈和联邦学习的多星协作频谱认知方法研究

doi:10.11959/j.issn.1000-436x.2024034

通信学报 ›› 2024, Vol. 45 ›› Issue (2): 90-105.doi: 10.11959/j.issn.1000-436x.2024034

• 学术论文 • 上一篇

融合Stackelberg博弈和联邦学习的多星协作频谱认知方法研究

丁晓进¹, 徐叶辉², 包文², 张更新¹

¹ 南京邮电大学通信与信息工程学院，江苏南京 210003
² 南京邮电大学物联网学院，江苏南京 210003

修回日期:2023-10-15 出版日期:2024-02-01 发布日期:2024-02-01
作者简介:丁晓进（1981− ），男，江苏兴化人，博士，南京邮电大学副教授，主要研究方向为卫星物联网、频谱智能认知等
徐叶辉（1999− ），男，江苏兴化人，南京邮电大学硕士生，主要研究方向为卫星通信
包文（1999− ），女，江苏扬州人，南京邮电大学硕士生，主要研究方向为卫星通信
张更新（1967− ），男，浙江平湖人，博士，南京邮电大学教授、博士生导师，主要研究方向为卫星通信、深空通信、空间信息网络等
基金资助:
国家自然科学基金资助项目(91738201);国家自然科学基金资助项目(62171234);2021-JCJQ-LB-006重点实验室基金资助项目(6142411422118)

Study on multi-satellite cooperative spectrum cognitive method integrating Stackelberg game and federated learning

Xiaojin DING¹, Yehui XU², Wen BAO², Gengxin ZHANG¹

¹ School of Communications and Information Engineering, Nanjing University of Posts and Telecommunications, Nanjing 210003, China
² School of Internet of Things, Nanjing University of Posts and Telecommunications, Nanjing 210003, China

Revised:2023-10-15 Online:2024-02-01 Published:2024-02-01
Supported by:
The National Natural Science Foundation of China(91738201);The National Natural Science Foundation of China(62171234);The National Key Laboratory Foundation of 2021-JCJQ-LB-006(6142411422118)

摘要/Abstract

摘要：

针对低成本和小型化低轨卫星的监测角度和方向分辨率相对较低、处理能力和峰值功率受限等因素造成单颗低轨卫星频谱认知能力弱的问题，提出了融合Stackelberg博弈和联邦学习的多星协作频谱认知方法。首先，结合各频谱认知卫星的可用算力、认知性能、处理与传输时延等特性，建立面向多频谱认知任务的协作卫星选择与算力资源分配算法；其次，基于所选择的节点和所分配的算力，设计低复杂度的多星协作频谱认知策略，其可自动辨识频谱空洞、检测干扰和识别调制模式。仿真实验结果表明，相比于单节点认知方法，所设计多星协作频谱认知策略能显著提升认知性能，且相比于对比模型，所设计策略中的模型可在不损失性能时，模型的参数量和浮点运算量降低分别可达96.69%和93.32%。

关键词: 卫星频谱认知, 多星协作, Stackelberg博弈, 联邦学习

Abstract:

To solve the problem of the weak spectrum-cognitive ability caused by monitoring angle, direction resolution, limited processing ability and peak power for a low-earth-orbit (LEO) satellite, a multi-satellite cooperative spectrum cognitive method integrating Stackelberg game and federated learning was proposed.Firstly, considering the available computing resource, cognitive performance, processing and transmission delay of each spectrum cognitive satellite, a cooperative-satellite selection and computing-resource allocation algorithm was built for multiple spectrum-cognitive tasks.Secondly, based on the selected satellites and the allocated computing resources, a low-complexity multi-satellite cooperative spectrum cognitive strategy was further designed, which could automatically sense the spectrum holes, and detect interference as well as identify the modulation mode.Simulation results demonstrate that compared to the single-node cognitive method, the designed multi-satellite cooperative spectrum cognitive strategy can obtain a better cognitive performance.Moreover, comparing with the existing model, the model utilized in the designed strategy can effectively achieve 96.69% and 93.32% lower number of parameters and required floating point operations per second, whilst maintaining the performance.

Key words: satellite spectrum cognitive, multi-satellite cooperation, Stackelberg game, federated learning

中图分类号:

TN92

丁晓进, 徐叶辉, 包文, 张更新. 融合Stackelberg博弈和联邦学习的多星协作频谱认知方法研究[J]. 通信学报, 2024, 45(2): 90-105.

Xiaojin DING, Yehui XU, Wen BAO, Gengxin ZHANG. Study on multi-satellite cooperative spectrum cognitive method integrating Stackelberg game and federated learning[J]. Journal on Communications, 2024, 45(2): 90-105.

图/表 23

图1

图2

图3

图4

图5

表1

图6

图7

图8

图9

图10

图11

图12

图13

表2

图14

图15

图16

表3

表4

表5

图17

表6

参考文献 22

[1]	YIN L , YANG R N , YANG Y Z ,et al. Beam pointing optimization based downlink interference mitigation technique between NGSO satellite systems[J]. IEEE Wireless Communications Letters, 2021,10(11): 2388-2392.
[2]	CHEN Q , GIAMBENE G , YANG L ,et al. Analysis of inter-satellite link paths for LEO mega-constellation networks[J]. IEEE Transactions on Vehicular Technology, 2021,70(3): 2743-2755.
[3]	KASSING S , BHATTACHERJEE D , áGUAS A B , et al . Exploring the “Internet from space” with Hypatia[C]// Proceedings of the ACM Internet Measurement Conference. New York:ACM Press, 2020: 1-16.
[4]	SU Y T , LIU Y Q , ZHOU Y Q ,et al. Broadband LEO satellite communications:architectures and key technologies[J]. IEEE Wireless Communications, 2019,26(2): 55-61.
[5]	CLARK B , GUNZINGER M . Winning the airwaves:regaining America’s dominance in the electromagnetic spectrum[R]. 2015.
[6]	ZHANG C , JIANG C X , JIN J ,et al. Spectrum sensing and recognition in satellite systems[J]. IEEE Transactions on Vehicular Technology, 2019,68(3): 2502-2516.
[7]	DING X J , NI T , ZOU Y L ,et al. Deep learning for satellites based spectrum sensing systems:a low computational complexity perspective[J]. IEEE Transactions on Vehicular Technology, 2023,72(1): 1366-1371.
[8]	王嘉煜, 陈思宏, 张骞丹 . 浅析一种基于DTP的卫星载波监视流程及方法[J]. 中国设备工程, 2022(12): 3-6.
	WANG J Y , CHEN S H , ZHANG Q D . Analysis of a satellite carrier monitoring process and method based on DTP[J]. China Plant Engineering, 2022(12): 3-6.
[9]	JIA M , LIU X , YIN Z S ,et al. Joint cooperative spectrum sensing and spectrum opportunity for satellite cluster communication networks[J]. Ad Hoc Networks, 2017,58: 231-238.
[10]	WANG Y F , DING X J , HONG T ,et al. Distributed-satellite-clusters-based spectrum sensing with two-stage phase alignment[J]. Sensors, 2022,22(11): 3983.
[11]	黄旭民, 张旸, 余荣 ,等. 基于 Stackelberg 博弈的无人机辅助无线供能物联网能量优化[J]. 通信学报, 2022,43(12): 146-156.
	HUANG X M , ZHANG Y , YU R ,et al. Stackelberg game based energy optimization for unmanned aerial vehicle assisted wireless-powered Internet of things[J]. Journal on Communications, 2022,43(12): 146-156.
[12]	CHEN Z B , XU Y Q , WANG H B ,et al. Federated learning-based cooperative spectrum sensing in cognitive radio[J]. IEEE Communications Letters, 2022,26(2): 330-334.
[13]	SUN W , LIAN S , ZHANG H ,et al. Lightweight digital twin and federated learning with distributed incentive in air-ground 6G networks[J]. IEEE Transactions on Network Science and Engineering, 2023,10(3): 1214-1227.
[14]	YANG Z H , CHEN M Z , SAAD W ,et al. Energy efficient federated learning over wireless communication networks[J]. IEEE Transactions on Wireless Communications, 2021,20(3): 1935-1949.
[15]	ACHTERBERG T . SCIP:solving constraint integer programs[J]. Mathematical Programming Computation, 2009,1(1): 1-41.
[16]	CIBLAT P , GHOGHO M . Blind NLLS carrier frequency-offset estimation for QAM,PSK and PAM modulations:performance at low SNR[J]. IEEE Transactions on Communications, 2006,54(10): 1725-1730.
[17]	AN Z L , ZHANG T Q , SHEN M ,et al. Series-constellation feature based blind modulation recognition for beyond 5G MIMO-OFDM systems with channel fading[J]. IEEE Transactions on Cognitive Communications and Networking, 2022,8(2): 793-811.
[18]	任杰, 高岭, 于佳龙 ,等. 面向边缘设备的高能效深度学习任务调度策略[J]. 计算机学报, 2020,43(3): 440-452.
	REN J , GAO L , YU J L ,et al. Energy-efficient deep learning task scheduling strategy for edge device[J]. Chinese Journal of Computers, 2020,43(3): 440-452.
[19]	CHEN Z B , XU Y Q , WANG H B ,et al. Deep STFT-CNN for spectrum sensing in cognitive radio[J]. IEEE Communications Letters, 2021,25(3): 864-868.
[20]	张帆, 黄赟, 方子茁 ,等. 卷积神经网络的损失最小训练后参数量化方法[J]. 通信学报, 2022,43(4): 114-122.
	ZHANG F , HUANG Y , FANG Z Z ,et al. Lost-minimum post-training parameter quantization method for convolutional neural network[J]. Journal on Communications, 2022,43(4): 114-122.
[21]	LUO B , LI X , WANG S Q ,et al. Cost-effective federated learning in mobile edge networks[J]. IEEE Journal on Selected Areas in Communications, 2021,39(12): 3606-3621.
[22]	YANG T T , NING J H , LAN D P ,et al. Kubeedge wireless for integrated communication and computing services everywhere[J]. IEEE Wireless Communications, 2022,29(2): 140-145.

层名称	输出大小	池化尺寸	步长	堆叠次数	输出通道数
频谱图像	224× 224	—	—	—	3
Conv1MaxPool	112×11256×56	3×33×3	22	1	24
Stage2	28×2828×28	—	21	11	36
Stage3	14×1414×14	—	21	13	48
Stage4	7×77×7	—	21	11	96
Conv5	7×7	1×1	1	1	190
GlobalPool	1×1	7×7	—	—
FC+Softmax	—	—	—	—	N×1

剪枝率	正确率
0	81.14%
5%	81.09%
10%	80.95%
19%	73.15%
22%	72.51%
25%	70.61%
30%	66.97%

卫星数量	正确率	能效
1	83.44%	0.73%
3	85.26%	0.25%
5	91.47%	0.16%
7	91.86%	0.09%

模型	参数量	FLOPs
ShuffleNetV2	1 256 679	151 687 864
ShuffleNetV2-RCPQ	41 499	10 135 344

模型	频谱感知时间/s	干扰检测时间/s	调制模式时间/s	总时间/s
ShuffleNetV2（结合基	27.228 8	43.374 6	62.167	132.770 4
于Stackelberg的博弈的
分配算法）
ShuffleNetV2-RCPQ	11.194	17.760 6	28.990	57.944 6
（结合基于Stackelberg
的博弈的分配算法）
ShuffleNetV2-RCPQ	10.872	17.152 2	32.742	60.766 2
（平均分配）

融合Stackelberg博弈和联邦学习的多星协作频谱认知方法研究

Study on multi-satellite cooperative spectrum cognitive method integrating Stackelberg game and federated learning

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 22

相关文章 15

Metrics

推荐阅读 0

评估方式	频谱感知性能（P_d）	干扰检测性能（正确率）	调制模式性能（正确率）
测试	90.63%	83.69%	85.51%
仿真	90.85%	84.81%	86.60%

[1]	吴昊, 郝佳佳, 卢云龙. 物联网场景下基于蜜场的分布式网络入侵检测系统研究[J]. 通信学报, 2024, 45(1): 106-118.
[2]	陈晓霖, 昝道广, 吴炳潮, 关贝, 王永吉. 面向纵向联邦学习的对抗样本生成算法[J]. 通信学报, 2023, 44(8): 1-13.
[3]	马卓, 金嘉玉, 杨易龙, 刘洋, 应作斌, 李腾, 张俊伟. 基于门限同态加密的自适应联邦学习安全聚合方案[J]. 通信学报, 2023, 44(7): 76-85.
[4]	马鑫迪, 李清华, 姜奇, 马卓, 高胜, 田有亮, 马建峰. 面向Non-IID数据的拜占庭鲁棒联邦学习[J]. 通信学报, 2023, 44(6): 138-153.
[5]	金彪, 李逸康, 姚志强, 陈瑜霖, 熊金波. GenFedRL：面向深度强化学习智能体的通用联邦强化学习框架[J]. 通信学报, 2023, 44(6): 183-197.
[6]	王再见, 谷慧敏. 基于联合优化的网络切片资源分配策略[J]. 通信学报, 2023, 44(5): 234-245.
[7]	张佳乐, 朱诚诚, 孙小兵, 陈兵. 基于GAN的联邦学习成员推理攻击与防御方法[J]. 通信学报, 2023, 44(5): 193-205.
[8]	田有亮, 吴柿红, 李沓, 王林冬, 周骅. 基于激励机制的联邦学习优化算法[J]. 通信学报, 2023, 44(5): 169-180.
[9]	姜慧, 何天流, 刘敏, 孙胜, 王煜炜. 面向异构流式数据的高性能联邦持续学习算法[J]. 通信学报, 2023, 44(5): 123-136.
[10]	余晟兴, 陈泽凯, 陈钟, 刘西蒙. DAGUARD：联邦学习下的分布式后门攻击防御方案[J]. 通信学报, 2023, 44(5): 110-122.
[11]	李开菊, 许强, 王豪. 冗余数据去除的联邦学习高效通信方法[J]. 通信学报, 2023, 44(5): 79-93.
[12]	王冬, 秦倩倩, 郭开天, 刘容轲, 颜伟鹏, 任一支, 罗清彩, 申延召. 联邦学习中的模型逆向攻防研究综述[J]. 通信学报, 2023, 44(11): 94-109.
[13]	马千飘, 贾庆民, 刘建春, 徐宏力, 谢人超, 黄韬. 异构边缘计算环境下异步联邦学习的节点分组与分时调度策略[J]. 通信学报, 2023, 44(11): 79-93.
[14]	余晟兴, 陈钟. 基于同态加密的高效安全联邦学习聚合框架[J]. 通信学报, 2023, 44(1): 14-28.
[15]	汤凌韬, 王迪, 刘盛云. 面向非独立同分布数据的联邦学习数据增强方案[J]. 通信学报, 2023, 44(1): 164-176.