Wireless traffic prediction based on federated learning

doi:10.11772/j.issn.1001-9081.2022050721

Journal of Computer Applications ›› 2023, Vol. 43 ›› Issue (6): 1900-1909.DOI: 10.11772/j.issn.1001-9081.2022050721

Special Issue: 网络与通信

• Network and communications • Previous Articles Next Articles

Wireless traffic prediction based on federated learning

Shangjing LIN¹(), Ji MA², Bei ZHUANG¹, Yueying LI¹, Ziyi LI¹, Tie LI¹, Jin TIAN²

^1.Beijing Key Laboratory of Work Safety Intelligent Monitoring （Beijing University of Posts and Telecommunications），Beijing 100876，China
^2.School of Networking and Communication Engineering，Jinling Institute of Technology，Nanjing Jiangsu 211169，China

Received:2022-05-20 Revised:2022-07-29 Accepted:2022-08-04 Online:2023-06-08 Published:2023-06-10
Contact: Shangjing LIN
About author:MA Ji， born in 1982， Ph. D.， lecturer. His research interests include deep learning， distributed algorithm， edge computing.
ZHUANG Bei， born in 2000， M. S. candidate. Her research interests include big data， edge computing.
LI Yueying， born in 1998， M. S. candidate. Her research interests include big data， edge computing.
LI Ziyi， born in 2000， M. S. candidate. Her research interests include wireless communication network.
LI Tie， born in 1999， M. S. candidate. His research interests include wireless communication network.
TIAN Jin， born in 1963， Ph. D.， professor. His research interests include artificial intelligence algorithm， application of Markov theory， wireless network media access control layer protocol， intelligent traffic system.
Supported by:
National Key Research and Development Program of China(2019YFC1511400);Fundamental Research Funds for the Central Universities （BUPT）(2021RC07);Open Fund for Key Laboratory of Universal Wireless Communication, Ministry of Education(KFKT-2020102);BUPT Postgraduate Innovation and Entrepreneurship Program(2022-YC-A086)

基于联邦学习的无线通信流量预测

林尚静¹(), 马冀², 庄琲¹, 李月颖¹, 李子怡¹, 李铁¹, 田锦²

^1.安全生产智能监控北京市重点实验室（北京邮电大学），北京 100876
^2.金陵科技学院网络与通信工程学院，南京 211169

通讯作者: 林尚静
作者简介:林尚静（1986—），女，湖北武汉人，讲师，博士，主要研究方向：深度学习、分布式算法、边缘计算Email：linshangjing@bupt.edu.cn
马冀（1982—），男，河北石家庄人，讲师，博士，主要研究方向：深度学习、分布式算法、边缘计算
庄琲（2000—），女，山东菏泽人，硕士研究生，主要研究方向：大数据、边缘计算
李月颖（1998—），女，江苏南通人，硕士研究生，主要研究方向：大数据、边缘计算
李子怡（2000—），女，河北保定人，硕士研究生，主要研究方向：无线通信网络
李铁（1999—），男，河北邯郸人，硕士研究生，主要研究方向：无线通信网络
田锦（1963—），男，四川宜宾人，教授，博士，主要研究方向：人工智能算法、马尔可夫理论应用、无线网络媒体接入控制层协议、智能交通系统。
基金资助:
国家重点研发计划项目(2019YFC1511400);北京邮电大学中央高校基本科研业务费新进教师人才项目(2021RC07);泛网无线通信教育部重点实验室开放基金资助项目(KFKT-2020102);北京邮电大学研究生创新创业资助项目(2022-YC-A086)

Abstract

Abstract:

Wireless communication network traffic prediction is of great significance to operators in network construction， base station wireless resource management and user experience improvement. However， the existing centralized algorithm models face the problems of complexity and timeliness， so that it is difficult to meet the traffic prediction requirements of the whole city scale. Therefore， a distributed wireless traffic prediction framework under cloud-edge collaboration was proposed to realize traffic prediction based on single grid base station with low complexity and communication overhead. Based on the distributed architecture， a wireless traffic prediction model based on federated learning was proposed. Each grid traffic prediction model was trained synchronously， JS （Jensen-Shannon） divergence was used to select grid traffic models with similar traffic distributions through the center cloud server， and Federated Averaging （FedAvg） algorithm was used to fuse the parameters of the grid traffic models with similar traffic distributions， so as to improve the model generalization and describe the regional traffic accurately at the same time. In addition， as the traffic in different areas within the city was highly differentiated in features， on the basis of the algorithm， a federated training method based on coalitional game was proposed. Combined with super-additivity criteria， the grids were taken as participants in the coalitional game， and screened. And the core of the coalitional game and the Shapley value were introduced for profit distribution to ensure the stability of the alliance， thereby improving the accuracy of model prediction. Experimental results show that taking Short Message Service （SMS） traffic as an example， compared with grid-independent training， the proposed model has the prediction error decreased most significantly in the suburb， with a decline range of 26.1% to 28.7%， the decline range is 0.7% to 3.4% in the urban area， and 0.8% to 4.7% in the downtown area. Compared with the grid-centralized training， the proposed model has the prediction error in the three regions decreased by 49.8% to 79.1%.

Key words: federated learning, cloud-edge collaboration, wireless traffic prediction, JS (Jensen-Shannon) divergence, coalitional game

摘要：

无线通信网络流量预测对运营商建设网络、管理基站无线资源和提升用户体验具有重要意义。然而，现有的集中式算法模型面临着复杂性和时效性问题，难以满足城市全域尺度的通信流量预测需求。因此，提出一个分布式的云边协同下的无线通信流量预测框架，以较低的复杂度和通信开销实现基于单栅格基站的流量预测。在分布式架构的基础上，提出了基于联邦学习的无线通信流量预测模型。各个栅格流量预测模型同步训练，通过中心云服务器利用JS（Jensen-Shannon）散度挑选出流量分布相似的栅格流量模型，并采用联邦平均（FedAvg）算法融合具有相似流量分布的栅格流量模型的参数，从而在提高模型泛化性的同时达到保持对本地流量精准刻画的目的。此外，由于城市范围内不同地区流量具有高度差异化的特征，在FedAvg的基础上，提出了基于合作博弈的联邦训练方法，将栅格作为合作博弈的参与者，结合超可加性准则筛选栅格，并引入合作博弈的核和夏普利值分配收益以确保联盟的稳定性，提高模型预测的准确性。实验结果表明，以短消息业务（SMS）流量为例，与栅格独立式训练相比，所提模型的预测误差下降在郊区最为明显，下降范围在26.1%~28.7%，在市区下降范围在0.7%~3.4%，在市中心下降范围在0.8%~4.7%；与栅格集中式训练相比，3个区域的模型预测误差下降范围在49.8%~79.1%。

关键词: 联邦学习, 云边协同, 无线流量预测, JS散度, 合作博弈

CLC Number:

TN92

Shangjing LIN, Ji MA, Bei ZHUANG, Yueying LI, Ziyi LI, Tie LI, Jin TIAN. Wireless traffic prediction based on federated learning[J]. Journal of Computer Applications, 2023, 43(6): 1900-1909.

林尚静, 马冀, 庄琲, 李月颖, 李子怡, 李铁, 田锦. 基于联邦学习的无线通信流量预测[J]. 《计算机应用》唯一官方网站, 2023, 43(6): 1900-1909.

Figures/Tables 16

References 18

1	刘少杰. 基于大数据的移动网络流量预测研究［D］. 北京：北方工业大学， 2021：31-41. 10.51393/j.jamst.2022015
	LIU S J. Research on mobile network traffic prediction based on big data［D］. Beijing： North China University of Technology， 2021：31-41. 10.51393/j.jamst.2022015
2	韩超，宋苏，王成红. 基于ARIMA模型的短时交通流实时自适应预测［J］. 系统仿真学报， 2004， 16（7）：1530-1532， 1535. 10.3969/j.issn.1004-731X.2004.07.042
	HAN C， SONG S， WANG C H. A real-time short-term traffic flow adaptive forecasting method based on ARIMA model ［J］. Journal of System Simulation， 2004， 16（7）：1530-1532， 1535. 10.3969/j.issn.1004-731X.2004.07.042
3	高波，张钦宇，梁永生，等. 基于EMD及ARMA的自相似网络流量预测［J］. 通信学报， 2011， 32（4）：47-56. 10.3969/j.issn.1000-436X.2011.04.007
	GAO B， ZHANG Q Y， LIANG Y S， et al. Predicting self-similar networking traffic based on EMD and ARMA ［J］. Journal on Communications， 2011， 32（4）：47-56. 10.3969/j.issn.1000-436X.2011.04.007
4	盛虎，张玉雪. 基于ARIMA的网络流量建模及预测研究［J］. 通信技术， 2019， 52（4）：903-907. 10.3969/j.issn.1002-0802.2019.04.022
	SHENG H， ZHANG Y X. Network traffic modeling and forecasting based on ARIMA［J］. Communications Technology， 2019， 52（4）：903-907. 10.3969/j.issn.1002-0802.2019.04.022
5	胡建伟. 基于LSTM的移动通信基站流量预测［J］. 信息技术与信息化， 2021（5）：84-86. 10.3969/j.issn.1672-9528.2021.05.024
	HU J W. Traffic prediction of mobile communication base station based on LSTM［J］. Information Technology and Informatization， 2021（5）：84-86. 10.3969/j.issn.1672-9528.2021.05.024
6	李校林，吴腾. 基于PF-LSTM网络的高效网络流量预测方法［J］. 计算机应用研究， 2019， 36（12）：3833-3836. 10.19734/j.issn.1001-3695.2018.06.0394
	LI X L， WU T. Efficient network traffic prediction method based on PF-LSTM network ［J］. Application Research of Computers， 2019， 36（12）：3833-3836. 10.19734/j.issn.1001-3695.2018.06.0394
7	TIAN Z D. Network traffic prediction method based on wavelet transform and multiple models fusion ［J］. International Journal of Communication Systems， 2020， 33（11）： No.e4415. 10.1002/dac.4415
8	康梦轩，宋俊平，范鹏飞，等. 基于深度学习的网络流量预测研究综述［J］. 计算机工程与应用， 2021， 57（10）：1-9. 10.3778/j.issn.1002-8331.2101-0402
	KANG M X， SONG J P， FAN P F， et al. Survey of network traffic forecast based on deep learning［J］. Computer Engineering and Applications， 2021， 57（10）：1-9. 10.3778/j.issn.1002-8331.2101-0402
9	McMAHAN H B， MOORE E， RAMAGE D， et al. Communication-efficient learning of deep networks from decentralized data［C］// Proceedings of the 20th International Conference on Artificial Intelligence and Statistics. New York： JMLR.org， 2016：1273-1282.
10	ZHU H Y， ZHANG H Y， JIN Y C. From federated learning to federated neural architecture search： a survey ［J］. Complex and Intelligent Systems， 2021， 7（2）：639-657. 10.1007/s40747-020-00247-z
11	YANG Q， LIU Y， CHEN T J， et al. Federated machine learning： concept and applications［J］. ACM Transactions on Intelligent Systems and Technology， 2019， 10（2）： No.12. 10.1145/3298981
12	MA Z Z， ZHAO M Y， CAI X J， et al. Fast-convergent federated learning with class-weighted aggregation ［J］. Journal of Systems Architecture， 2021， 117： No.102125. 10.1016/j.sysarc.2021.102125
13	BALAKRISHNAN R， AKDENIZ M， DHAKAL S， et al. Resource management and model personalization for federated learning over wireless edge networks［J］. Journal of Sensor and Actuator Networks， 2021， 10（1）： No.17. 10.3390/jsan10010017
14	DINH C T， TRAN N H， NGUYEN T D. Personalized federated learning with Moreau envelopes［C］// Proceedings of the 34th International Conference on Neural Information Processing Systems. Red Hook， NY： Curran Associates Inc.， 2020：21394-21405. 10.1145/3404397.3404457
15	SAAD W， HAN Z， DEBBAH M， et al. Coalitional game theory for communication networks： a tutorial［J］. IEEE Signal Processing Magazine， 2009， 26（5）：77-97. 10.1109/msp.2009.000000
16	谢娟英，丁丽娟，王明钊. 基于谱聚类的无监督特征选择算法［J］. 软件学报， 2020， 31（4）：1009-1024. 10.13328/j.cnki.jos.005927
	XIE J Y， DING L J， WANG M Z. Spectral clustering based unsupervised feature selection algorithms ［J］. Journal of Software， 2020， 31（4）： 1009-1024. 10.13328/j.cnki.jos.005927
17	GREFF K， SRIVASTAVA R K， KOUTNÍK J， et al. LSTM： a search space odyssey ［J］. IEEE Transactions on Neural Networks and Learning Systems， 2017， 28（10）： 2222-2232. 10.1109/tnnls.2016.2582924
18	AUMANN R. The Shapley value ［M］// ICHIISHI T， NEYMAN A， TAUMAN Y. Game Theory and Applications： A Volume in Economic Theory， Econometrics， and Mathematical Economics. Cambridge： Academic Press， 1990：158-165. 10.1016/b978-0-12-370182-4.50011-8

模型	优缺点
本文模型	云边协同的城市流量预测框架能够以较低的复杂度、较高的实时性实现城市全域流量预测；基于合作博弈的联邦训练模型针对Non-IID的通信流量数据提出模型优化方法，有效提高通信流量预测的精确性
自回归滑动平均模型^［3］	模型简单，计算复杂度低，可以准确刻画自相似网络流量；但对于存在网络异常的长时间网络流量数据并不适用
自回归移动平均模型^［4］	模型简单，计算复杂度低；但预测精度低，仅适用于短期流量预测
PF-LSTM模型^［6］	克服了传统LSTM网络收敛于局部最优的缺点；但训练流程相对复杂

模型	优缺点
本文模型	云边协同的城市流量预测框架能够以较低的复杂度、较高的实时性实现城市全域流量预测；基于合作博弈的联邦训练模型针对Non-IID的通信流量数据提出模型优化方法，有效提高通信流量预测的精确性
自回归滑动平均模型^［3］	模型简单，计算复杂度低，可以准确刻画自相似网络流量；但对于存在网络异常的长时间网络流量数据并不适用
自回归移动平均模型^［4］	模型简单，计算复杂度低；但预测精度低，仅适用于短期流量预测
PF-LSTM模型^［6］	克服了传统LSTM网络收敛于局部最优的缺点；但训练流程相对复杂

指标	模型	市区			市中心			郊区
指标	模型	栅格1	栅格2	栅格3	栅格1	栅格2	栅格3	栅格1	栅格2	栅格3
MSE	非联邦学习	0.795	67.098	68.850	176.013	157.220	192.501	1.628	1.657	1.657
	集中式学习	2.757	18.158	9.666	505.772	811.141	848.464	3.524	3.255	19.358
	联邦学习	0.783	64.990	64.286	159.865	145.804	186.982	0.889	0.843	0.846
RMSE	非联邦学习	0.891	8.191	8.298	13.267	12.539	13.874	1.276	1.287	1.287
	集中式学习	1.660	4.261	3.109	22.489	28.481	29.128	1.877	1.804	4.400
	联邦学习	0.885	8.062	8.018	12.644	12.075	13.764	0.943	0.918	0.920
R²	非联邦学习	0.354	0.622	0.608	0.607	0.610	0.596	0.531	0.523	0.524
	集中式学习	0.689	0.377	0.387	0.811	0.763	0.549	0.507	0.598	0.525
	联邦学习	0.399	0.588	0.589	0.629	0.638	0.614	0.493	0.508	0.512

指标	模型	市区			市中心			郊区
指标	模型	栅格1	栅格2	栅格3	栅格1	栅格2	栅格3	栅格1	栅格2	栅格3
MSE	非联邦学习	0.795	67.098	68.850	176.013	157.220	192.501	1.628	1.657	1.657
	集中式学习	2.757	18.158	9.666	505.772	811.141	848.464	3.524	3.255	19.358
	联邦学习	0.783	64.990	64.286	159.865	145.804	186.982	0.889	0.843	0.846
RMSE	非联邦学习	0.891	8.191	8.298	13.267	12.539	13.874	1.276	1.287	1.287
	集中式学习	1.660	4.261	3.109	22.489	28.481	29.128	1.877	1.804	4.400
	联邦学习	0.885	8.062	8.018	12.644	12.075	13.764	0.943	0.918	0.920
R²	非联邦学习	0.354	0.622	0.608	0.607	0.610	0.596	0.531	0.523	0.524
	集中式学习	0.689	0.377	0.387	0.811	0.763	0.549	0.507	0.598	0.525
	联邦学习	0.399	0.588	0.589	0.629	0.638	0.614	0.493	0.508	0.512

指标	模型	SMS			Voice			Internet
指标	模型	市区栅格	市中心栅格	郊区栅格	市区栅格	市中心栅格	郊区栅格	市区栅格	市中心栅格	郊区栅格
MSE	非联邦学习	0.795	176.013	1.628	0.314	320.364	0.684	12.122	130 796.234	238.509
	集中式学习	2.757	505.772	3.524	0.882	938.018	1.282	43.396	16 142.886	260.976
	联邦学习	0.783	159.865	0.889	0.302	270.755	0.678	2.836	13 201.060	100.583
RMSE	非联邦学习	0.891	13.267	1.276	0.560	17.899	0.827	3.482	361.658	15.444
	集中式学习	1.660	22.489	1.877	0.939	30.627	1.132	6.588	127.055	16.155
	联邦学习	0.885	12.644	0.943	0.550	16.455	0.823	1.684	114.896	10.029
R²	非联邦学习	0.354	0.607	0.531	0.468	0.625	0.592	0.597	0.586	0.590
	集中式学习	0.689	0.811	0.507	0.762	0.699	0.514	-1.674	0.521	0.635
	联邦学习	0.399	0.629	0.493	0.510	0.233	0.596	0.825	0.608	0.859

Wireless traffic prediction based on federated learning

基于联邦学习的无线通信流量预测

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 16

References 18

Related Articles 15

Recommended Articles

Metrics

[1]	Tingwei CHEN, Jiacheng ZHANG, Junlu WANG. Random validation blockchain construction for federated learning [J]. Journal of Computer Applications, 2024, 44(9): 2770-2776.
[2]	Zheyuan SHEN, Keke YANG, Jing LI. Personalized federated learning method based on dual stream neural network [J]. Journal of Computer Applications, 2024, 44(8): 2319-2325.
[3]	Wei LUO, Jinquan LIU, Zheng ZHANG. Dual vertical federated learning framework incorporating secret sharing technology [J]. Journal of Computer Applications, 2024, 44(6): 1872-1879.
[4]	Xuebin CHEN, Zhiqiang REN, Hongyang ZHANG. Review on security threats and defense measures in federated learning [J]. Journal of Computer Applications, 2024, 44(6): 1663-1672.
[5]	Sunjie YU, Hui ZENG, Shiyu XIONG, Hongzhou SHI. Incentive mechanism for federated learning based on generative adversarial network [J]. Journal of Computer Applications, 2024, 44(2): 344-352.
[6]	Zucuan ZHANG, Xuebin CHEN, Rui GAO, Yuanhuai ZOU. Federated learning client selection method based on label classification [J]. Journal of Computer Applications, 2024, 44(12): 3759-3765.
[7]	Xuebin CHEN, Changsheng QU. Overview of backdoor attacks and defense in federated learning [J]. Journal of Computer Applications, 2024, 44(11): 3459-3469.
[8]	Shuaihua ZHANG, Shufen ZHANG, Mingchuan ZHOU, Chao XU, Xuebin CHEN. Malicious traffic detection model based on semi-supervised federated learning [J]. Journal of Computer Applications, 2024, 44(11): 3487-3494.
[9]	Jie WU, Xuezhong QIAN, Wei SONG. Personalized federated learning based on similarity clustering and regularization [J]. Journal of Computer Applications, 2024, 44(11): 3345-3353.
[10]	Chunyong YIN, Yongcheng ZHOU. Automatically adjusted clustered federated learning for double-ended clustering [J]. Journal of Computer Applications, 2024, 44(10): 3011-3020.
[11]	Hui ZHOU, Yuling CHEN, Xuewei WANG, Yangwen ZHANG, Jianjiang HE. Deep shadow defense scheme of federated learning based on generative adversarial network [J]. Journal of Computer Applications, 2024, 44(1): 223-232.
[12]	Mengjie LAN, Jianping CAI, Lan SUN. Self-regularization optimization methods for Non-IID data in federated learning [J]. Journal of Computer Applications, 2023, 43(7): 2073-2081.
[13]	Wanzhen CHEN, En ZHANG, Leiyong QIN, Shuangxi HONG. Privacy-preserving federated learning algorithm based on blockchain in edge computing [J]. Journal of Computer Applications, 2023, 43(7): 2209-2216.
[14]	Sheng YE, Jing WANG, Jianfeng XIN, Guiling WANG, Chenhong GUO. Dynamic evolution method for microservice composition systems in cloud-edge environment [J]. Journal of Computer Applications, 2023, 43(6): 1696-1704.
[15]	Shaochen HAO, Zizuan WEI, Yao MA, Dan YU, Yongle CHEN. Network intrusion detection model based on efficient federated learning algorithm [J]. Journal of Computer Applications, 2023, 43(4): 1169-1175.