空天地协同的边缘云服务功能链动态编排方法

doi:10.19665/j.issn1001-2400.2022.02.010

摘要/Abstract

摘要：

空天地协同网络作为一种新型网络架构,拥有广域网络覆盖和泛在无缝接入等优势,但也面临日益增长的用户需求与有限的网络服务资源之间的矛盾,将边缘计算引入空天地协同网络能够大幅提升系统的业务处理能力。同时,为了进一步提高空天地协同的边缘云网络资源利用率,可以将一组虚拟网络功能按照一定的业务逻辑连接起来,形成动态可重构的服务功能链,为用户提供多样化且高质量的网络服务。考虑到空天地协同的边缘云网络的高动态和异构特性,研究了动态服务功能链的高效编排问题,设计了空天地协同的边缘云网络系统模型,并在此基础上构造了受网络资源和服务请求约束的端到端时延优化目标函数。其次,结合量子机器学习高效并行计算的优势,将服务功能链路径选择问题建模为基于开放量子随机行走的隐马尔可夫模型,并采用量子回溯解码的方法对模型进行了求解。最后,对文章研究的内容进行仿真验证,仿真结果表明:与传统精确求解和启发式方法相比,文章中提出的服务功能链动态编排方法能够在网络流量负载较高的情况下,提高服务请求成功率并降低端到端平均时延。

关键词: 空天地协同网络, 边缘云网络, 服务功能链, 量子机器学习

Abstract:

As a new network architecture,the space air terrestrial integrated network has the advantages of wide network coverage and ubiquitous seamless access ability,but it also faces the contradiction between increasing users’ demands and limited network service resources.Introducing edge computing into space air terrestrial integrated networks can greatly improve the system’s business processing capability.Meanwhile,first,in order to improve the resources utilization of Space Air Terrestrial aided Mobile-access Edge Cloud (SAT-MEC) networks,and provide users with diversified and high-quality network services,we connect a group of virtual network functions according to a certain business logic,which forms a dynamic and reconfigurable service function chain.Considering the high dynamic and heterogeneous characteristics of the SAT-MEC network,the efficient scheduling method for its dynamic service function chain is studied,and the system model of the SAT-MEC network is designed.On this basis,the objective function of end-to-end delay optimization constrained by network resources and service requests is constructed.Second,combining the advantages of efficient parallel computing of Quantum Machine Learning,the path selection problem of the service function chain is modeled as a Hidden Markov Model based on Open Quantum Random Walk,with the model solved by the Quantum Backtrack Decoding method.Compared with the traditional precise solution and heuristic methods,simulation results show that the proposed method can improve the success rate of service request and reduce the end-to-end average delay under the condition of a high network traffic load.

Key words: space air terrestrial network, edge cloud network, SDN/NFV, quantum machine learning

中图分类号:

TP393

乔文欣,卢昱,刘益岑,李志伟,李玺. 空天地协同的边缘云服务功能链动态编排方法[J]. 西安电子科技大学学报, 2022, 49(2): 79-88.

QIAO Wenxin,LU Yu,LIU Yicen,LI Zhiwei,LI Xi. Dynamic scheduling method for service function chains in space air terrestrial aided edge cloud networks[J]. Journal of Xidian University, 2022, 49(2): 79-88.

图/表 8

图1

表1

主要符号及描述"

符号	描述
G=(N,L)	底层网络拓扑G包含节点集合N和链路集合L
N=N_S∪N_A∪N_T	节点集合N包含卫星节点集合N_S、无人平台节点集合N_A和地面节点集合N_T
P,R	SFC路径集合和服务请求集合
$T n i, v x p r o c e s s$	第x个VNF v_x在节点n_i上的处理时延
$T n i, n j t r a n s f e r$ , $T n i, n j d u r a t i o n$	节点n_i和n_j之间的链路传输时延和链路可持续时间
η(v_x),δ(l)	VNF v_x所需的处理能力和链路l所需的带宽容量
π(n_i),ρ(p)	节点n_i的容量和路径p的容量
R=(E,F,D,T_d)	服务请求R包含请求所属边缘云E、所需VNFs集合F、时延D和生命周期T_d
$X n i r, v x$	服务请求r所需的VNF v_x是否映射到节点n_i的布尔变量,1代表映射,0代表否
$Y n i, n j, p r, v x, v y$	服务请求r所需的VNFs v_x和v_y是否映射到节点n_i和n_j的布尔变量,1代表映射,0代表否
$H l, p n i, n j$	虚拟链路l被节点n_i和n_j之间的路径p遍历的布尔数组,1代表遍历,0代表否

表1

图2

表2

图3

图4

图5

图6

参考文献 20

[1]	FERRÚS R, KOUMARAS H, SALLENT O, et al. SDN/NFV-Enabled Satellite Communications Networks:Opportunities,Scenarios and Challenges[J]. Physical Communication, 2016, 18:95-112. doi: 10.1016/j.phycom.2015.10.007
[2]	GIAMBENE G, KOTA S, PILLAI P. Satellite-5G Integration:A Network Perspective[J]. IEEE Network, 2018, 32(5):25-31. doi: 10.1109/MNET.2018.1800037
[3]	MARCHESE M, MOHEDDINE A, PATRONE F. IoT and UAV Integration in 5G Hybrid Terrestrial-Satellite Networks[J]. Sensors, 2019, 19(17):3704. doi: 10.3390/s19173704
[4]	LIU J, SHI Y, FADLULLAH Z M, et al. Space-Air-Ground Integrated Network:A Survey[J]. IEEE Communications Surveys & Tutorials, 2018, 20(4):2714-2741.
[5]	唐琴琴, 谢人超, 刘旭, 等. 融合MEC的星地协同网络:架构,关键技术与挑战[J]. 通信学报, 2020, 41(4):162-181. doi: 10.1111/j.1460-2466.1991.tb02315.x
	TANG Qinqin, XIE Renchao, LIU Xu, et al. MEC Enabled Satellite-Terrestrial Network:Architecture,Key Technique and Challenge[J]. Journal on Communications, 2020, 41(4):162-181. doi: 10.1111/j.1460-2466.1991.tb02315.x
[6]	刘益岑, 陈兴凯, 卢昱, 等. 一种软件定义网络的安全服务路径优化构建机制[J]. 西安电子科技大学学报, 2019, 46(1):158-165.
	LIU Yiceng, CHEN Xingkai, LU Yu, et al. SDN-Based Optimal Security Service Path Construction Mechanism[J]. Journal of Xidian University, 2019, 46(1):158-165.
[7]	EVANS B, WERNER M, LUTZ E, et al. Integration of Satellite and Terrestrial Systems in Future Multimedia Communications[J]. IEEE Wireless Communications, 2005, 12(5):72-80. doi: 10.1109/MWC.2005.1522108
[8]	KOTA S, GIAMBENE G, KIM S. SatelliteComponent of NGN:Integrated and Hybrid Networks[J]. International Journal of Satellite Communications & Networking, 2011, 29(3):191-208.
[9]	CHENG N, XU W, SHI W, et al. Air-Ground Integrated Mobile Edge Networks:Architecture,Challenges and Opportunities[J]. IEEE Communications Magazine, 2018, 56(8):26-32.
[10]	SHI Y, CAO Y, LIU J, et al. A Cross-Domain SDN Architecture for Multi-Layered Space-Terrestrial Integrated Networks[J]. IEEE Network, 2018, 33(1):29-35. doi: 10.1109/MNET.2018.1800191
[11]	WANG F, JIANG D, QI S, et al. A Dynamic Resource Scheduling Scheme in Edge Computing Satellite Networks[J]. Mobile Networks and Applications, 2021, 26:597-608. doi: 10.1007/s11036-019-01421-5
[12]	CAO S, WEI J, HAN H, et al. Space Edge Cloud Enabling Network Slicing for 5G Satellite Network[C]// Proceedings of the 2019 15th International Wireless Communications & Mobile Computing Conference (IWCMC).Piscataway:IEEE, 2019:24-28.
[13]	ZHANG N, ZHANG S, YANG P, et al. Software Defined Space-Air-Ground Integrated Vehicular Networks:Challenges and Solutions[J]. IEEE Communications Magazine, 2017, 55(7):101-109.
[14]	VARASTEH A, HOFMANN S, DERIC N, et al. Mobility-Aware Joint Service Placement and Routing in Space-Air-Ground Integrated Networks[C]// Proceedings of the 2019 IEEE International Conference on Communications (ICC).Piscataway:IEEE, 2019:20-24.
[15]	WANG G, ZHOU S, ZHANG S, et al. SFC-Based Service Provisioning for Reconfigurable Space-Air-Ground Integrated Networks[J]. IEEE Journal on Selected Areas in Communications, 2020, 38(7):1478-1489. doi: 10.1109/JSAC.2020.2986851
[16]	LI G, ZHOU H, FENG B, et al. Horizontal-Based Orchestration for Multi-Domain SFC in SDN/NFV-Enabled Satellite/Terrestrial Networks[J]. China Communication, 2018, 15(5):77-91.
[17]	黄一鸣, 雷航, 李晓瑜. 量子机器学习算法综述[J]. 计算机学报, 2018, 41(1):145-163.
	HUANG Yiming, LEI Hang, LI Xiaoyu. A Survey on Quantum Machine Learning[J]. Chinese Journal of Computers, 2018, 41(1):145-163.
[18]	江逸茗, 兰巨龙, 周慧琴. 网络虚拟化环境下的资源监控策略[J]. 电子与信息学报, 2014, 36(3):708-714.
	JIANG Yiming, LAN Julong, ZHOU Huiqin. Resource Monitoring Policy for Network Virtualization Environment[J]. Journal of Electronics and Information Technology, 2014, 36(3):708-714.
[19]	CHOLEWA M, GAWRON P, GŁOMB P, et al. QuantumHidden Markov Models Based on Transition Operation Matrices[J]. Quantum Information Processing, 2017, 16(4):101. doi: 10.1007/s11128-017-1544-8
[20]	LI T, ZHOU H, LUO H, et al. SERvICE:A Software Defined Framework for Integrated Space-Terrestrial Satellite Communication[J]. IEEE Transactions on Mobile Computing, 2018, 17(3):703-716. doi: 10.1109/TMC.2017.2732343

参数	数量	计算资源/Unit	带宽资源/(Mb·s^-1)	处理时延/ms
核心云节点	1	2 000	[800,1 000]	[0.1,0.3]
MEC节点	15	[200,400]	[400,500]	[1,3]
LEO-MEC节点	1	200	[200,300]	[2,5]
SFC请求		[1,5]	[1,5]