车载边缘计算研究综述

doi:10.11959/j.issn.1000-0801.2023262

摘要/Abstract

摘要：

随着计算密集和时延敏感型车辆应用的爆炸式增长，集中式云架构产生了高工作负载和任务时延阻塞。为了保证服务质量，车载边缘计算应运而生，这种计算模式将计算能力和存储资源，推移到离数据源更近的边缘服务器或边缘网关等边缘节点上，通过在边缘节点进行实时数据处理和决策，可以显著地减少数据传输时延。首先介绍了车载边缘计算的基本概念，接着对现有研究进行了梳理分类，最后讨论了对车载边缘计算的展望和未来研究方向。

关键词: 移动边缘计算, 车联网, 车载边缘计算, 任务卸载

Abstract:

With the explosive growth of computation-intensive and latency-sensitive vehicular applications, the centralized cloud architecture was faced with high workload and task latency congestion.In order to ensure service quality, vehicular edge computing emerged.The computing paradigm enabled the migration of computational power and storage resources to edge servers or edge gateways that were closer to the data source.By performing real-time data processing and decision-making at the edge nodes, significant reduction in data transmission latency was achieved.Firstly, the basic concept of vehicular edge computing was introduced, followed by a review and classification of existing research.Finally, prospects and future research directions for vehicular edge computing were discussed.

Key words: mobile edge computing, Internet of vehicles, vehicular edge computing, task offloading

中图分类号:

TP391

彭雪飞, 刘奥辉. 车载边缘计算研究综述[J]. 电信科学, 2023, 39(12): 19-28.

Xuefei PENG, Aohui LIU. A survey of vehicular edge computing[J]. Telecommunications Science, 2023, 39(12): 19-28.

图/表 5

图1

图2

图3

图4

表1

车载边缘计算工作总结"

文献	协作方式	边缘节点	评估指标	贡献
文献[20]	端云	—	时延、能耗	将优化问题公式化为无限域半马尔可夫决策过程，利用迭代算法寻找最优方案
文献[21]	端边	多服务器	时延，能耗	提出了将边缘节点的放置优化问题描述成一个二进制整数线性规划问题，满足了不同V2X应用的QoS要求
文献[22]	端边云	多服务器	时延、能耗	提出云计算中心和边缘计算的协同式计算任务卸载架构
文献[23]	端边	多服务器	时延	引入软件定义网络辅助路由和控制协议，将 IoV 系统划分为两个独立的层，提出了一种联合优化卸载决策、卸载率和资源分配的方法，以最小化系统平均时延
文献[26]	端边	多服务器	能耗	同时考虑了计算能量成本和缓存能量成本问题，提出了一种基于深度确定性策略梯度算法来解决该优化问题
文献[27]	端边	多服务器	时延、能耗	设计了一种新颖的以V2X应用数据为中心的框架，最小化应用程序的总执行时间和成本
文献[28]	端边	多服务器	服务率	设计了一个分布式的服务调度机制，通过融合不同服务器的资源，确保实时的业务需求，最大限度提升服务率
文献[30]	端边	多服务器	时延、能耗	提出了低复杂度启发式贪婪算法，该算法以计算时延和成本函数为优化目标来选择每项任务成本最低的服务器
文献[32]	端边	多服务器	时延、能耗	开发了一种基于粒子群的启发式算法来解决优化问题
文献[36]	端边	多服务器	时延、能耗	提出了基于DQN的深度强化学习策略。优化状态（state）中的成本函数
文献[37]	端边	单服务器	缓存成本	提出了一种具有优先级的任务卸载和资源分配方案，以优化功率分配，在此基础上设计了DDPG算法
文献[38]	端边	多服务器	时延、可靠性	提出了一种面向车联网边缘计算的多目标强化学习策略，用于有效地减少系统总成本，包括时延和可靠性。该策略通过结合通信和计算资源分配，实现了在不同任务条件下平均约50%的总体系统成本降低
文献[43]	端边	多服务器	边缘节点服务率	将任务卸载子问题建模为精确势博弈，并提出了一种多智能体分布式深度确定性策略梯度来实现纳什均衡
文献[44]	端边	多服务器	服务器效益、能耗	将资源分配问题转化为一场博弈游戏，并提出一种去中心化的算法寻找纳什均衡
文献[45]	端边	多服务器	时延、能耗	将多用户卸载决策问题表述为非合作博弈，并提出了一种基于机器学习技术的全分布式计算卸载算法来寻找纳什均衡点

表1

参考文献 52

[1]	World Health Organization (WHO). Number of registered vehicles[EB]. 2020.
[2]	ZHANG J , LETAIEF K B . Mobile edge intelligence and computing for the internet of vehicles[J]. Proceedings of the IEEE, 2019,108(2): 246-261.
[3]	陈山枝 . 蜂窝车联网（C-V2X）[M]. 北京: 人民邮电出版社, 2020.
	CHEN S Z . Cellular vehicle-to-everything (C-V2X)[M]. Beijing: Posts ＆ Telecom Press, 2020.
[4]	KIM S , KIM B J . Crash risk-based prioritization of basic safety message in DSRC[J]. IEEE Access, 2020,8(1): 211961-211972.
[5]	WANG X , MAO S , GONG M X . An overview of 3GPP cellular vehicle-to-everythingstandards[J]. GetMobile:Mobile Computing and Communications, 2017,21(3): 19-25.
[6]	GHAFOOR K Z , GUIZANI M , KONG L ,et al. Enabling efficient coexistence of DSRC and C-V2X in vehicular networks[J]. IEEE Wireless Communications, 2019,27(2): 134-140.
[7]	ZHANG Y , ZHANG J X . Design and optimization of cluster-based DSRC and C-V2X hybrid routing[J]. Applied Sciences, 2022,12(13): 6782-6788.
[8]	AHMAD I , NOOR R M , ZABA M R ,et al. A cooperative heterogeneous vehicular clustering mechanism for road traffic management[J]. International Journal of Parallel Programming, 2020,48(5): 870-889.
[9]	ULLAH A , YAQOOB S , IMRAN M ,et al. Emergency message dissemination schemes based on congestion avoidance in VANET and vehicular FoG computing[J]. IEEE Access, 2018,7(7): 1570-1585.
[10]	National Highway Traffic Safety Administration (NHTSA). Traffic safety and the 5.9 GHz spectrum[EB]. 2019.
[11]	HUANG X M , YU R , KANG J W ,et al. Exploring mobile edge computing for 5G-enabled software defined vehicular networks[J]. IEEE Wireless Communications, 2018,24(6): 55-63.
[12]	KONG X J , WANG K L , WANG S P ,et al. Real-time mask identification for COVID-19:an edge-computing-based deep learning framework[J]. IEEE Internet of Things Journal, 2021,8(21): 15929-15938.
[13]	DING Z G , NG D W K , SCHOBER R ,et al. Delay minimization for NOMA-MEC offloading[J]. IEEE Signal Processing Letters, 2018,25(12): 1875-1879.
[14]	ZHOU X K , YANG X , MA J H ,et al. Energy-efficient smart routing based on link correlation mining for wireless edge computing in IoT[J]. IEEE Internet of Things Journal, 2021,9(16): 14988-14997.
[15]	AHMAD I , NOOR R M , ALI I ,et al. Characterizing the role of vehicular cloud computing in road traffic management[J]. International Journal of Distributed Sensor Networks, 2017,13(5): 1-14.
[16]	吕品, 许嘉, 李陶深 ,等. 面向自动驾驶的边缘计算技术研究综述[J]. 通信学报, 2021,42(3): 190-208.
	LYU P , XU J , LI T S ,et al. Overview of research on edge computing technology for automatic driving[J]. Journal of Communications, 2021,42(3): 190-208.
[17]	LIU S S , TANG J , ZHANG Z ,et al. Computer architectures for autonomous driving[J]. Computer, 2017,50(8): 18-25.
[18]	WANG Y , YIN K . Study of overtaking method of intelligent vehicle under vehicle road coordination[J]. Journal of Physics:Conference Series, 2021,41(1): 1-6.
[19]	YAN B , FANG C , QIU H ,et al. Intelligent speed limit system for safe expressway driving in rainy and foggy weather based on Internet of things[J]. Journal of Shanghai Jiaotong University (Science), 2023,28(1): 10-19.
[20]	ZHENG K , MENG H , CHATZIMISIOS P ,et al. An SMDP-based resource allocation in vehicular cloud computing systems[J]. IEEE Transactions on Industrial Electronics, 2015,62(12): 7920-7928.
[21]	MOUBAYED A , SHAMI A , HEIDARI P ,et al. Edge-enabled V2X service placement for intelligent transportation systems[J]. IEEE Transactions on Mobile Computing, 2020,20(4): 1380-1392.
[22]	ZHANG K , MAO Y M , LENG S P ,et al. Optimal delay constrained offloading for vehicular edge computing networks[C]// Proceedings of 2017 IEEE International Conference on Communications (ICC). Piscataway:IEEE Press, 2017: 1-6.
[23]	LIN L , ZHANG L . Joint optimization of offloading and resource allocation for SDN-enabled IoV[J]. Wireless Communications and Mobile Computing, 2022 (1): 1-13.
[24]	CONG P , ZHOU J , LI L ,et al. A survey of hierarchical energy optimization for mobile edge computing:a perspective from end devices to the cloud[J]. ACM Computing Surveys (CSUR), 2020,53(2): 1-44.
[25]	臧金环, 李春玲 . 《新能源汽车产业发展规划(2021—2035年)》调整解读[J]. 汽车工艺师, 2021(1): 32-34.
	ZANG J H , LI C L . Adjustment and interpretation of New Energy Vehicle Industry Development Plan(2021-2035)[J]. Auto Manufacturing Engineer, 2021(1): 32-34.
[26]	KONG X J , DUAN G H , HOU M L ,et al. Deep reinforcement learning-based energy-efficient edge computing for Internet of vehicles[J]. IEEE Transactions on Industrial Informatics, 2022,18(9): 6308-6316.
[27]	WU L , ZHANG R , LI Q ,et al. A mobile edge computing-based applications execution framework for Internet of vehicles[J]. Frontiers of Computer Science, 2022,16(5): 1-11.
[28]	LIAO Y , QIAO X , YU Q ,et al. Intelligent dynamic service pricing strategy for multi-user vehicle-aided MEC networks[J]. Future Generation Computer Systems, 2021,114(1): 15-22.
[29]	刘逸 . 粒子群优化算法的改进及应用研究[D]. 西安:西安电子科技大学, 2013.
	LIU Y . Research on improvement and application of particle swarm optimization algorithm[D]. Xi’an:Xidian University, 2013.
[30]	BELOGAEV A , ELOKHIN A , KRASILOV A ,et al. Cost-effective V2X task offloading in MEC-assisted intelligent transportation systems[J]. IEEE Access, 2020(8): 169010-169023.
[31]	CHEN M H , DONG M , LIANG B . Resource sharing of a computing access point for multi-user mobile cloud offloading with delay constraints[J]. IEEE Transactions on Mobile Computing, 2018,17(12): 2868-2881.
[32]	HUYNH L N T , PHAM Q V , PHAM X Q ,et al. Efficient computation offloading in multi-tier multi-access edge computing systems:a particle swarm optimization approach[J]. Applied Sciences, 2019,10(1): 203-210.
[33]	BUI K H N , JUNG J J . ACO-based dynamic decision making for connected vehicles in IoT system[J]. IEEE Transactions on Industrial Informatics, 2019,15(10): 5648-5655.
[34]	SUN Y , GUO X , SONG J ,et al. Adaptive learning-based task offloading for vehicular edge computing systems[J]. IEEE Transactions on Vehicular Technology, 2019,68(4): 3061-3074.
[35]	SUN Y , GUO X , ZHOU S ,et al. Learning-based task offloading for vehicular cloud computing systems[C]// Proceedings of 2018 IEEE International Conference on Communications. Piscataway:IEEE Press, 2018: 1-7.
[36]	ZHANG L H , ZHOU W Q , XIA J J ,et al. DQN-based mobile edge computing for smart Internet of vehicle[J]. EURASIP Journal on Advances in Signal Processing, 2022(1): 1-16.
[37]	HAZARIKA B , SINGH K , BISWAS S ,et al. DRL-based resource allocation for computation offloading in IoV networks[J]. IEEE Transactions on Industrial Informatics, 2022,18(11): 8027-8038.
[38]	CUI Y , DU L , WANG H ,et al. Reinforcement learning for joint optimization of communication and computation in vehicular networks[J]. IEEE Transactions on Vehicular Technology, 2021,70(12): 13062-13072.
[39]	WANG K , WANG X , LIU X ,et al. Task offloading strategy based on reinforcement learning computing in edge computing architecture of Internet of vehicles[J]. IEEE Access, 2020,8(1): 173779-173789
[40]	RANADHEERA S , MAGHSUDI S , HOSSAIN E . Mobile edge computation offloading using game theory and reinforcement learning[J]. 2017:arXiv:10.48550/arXiv.1711.09012.
[41]	MOURA J , HUTCHISON D . Game theory for multi-access edge computing:survey,use cases,and future trends[J]. IEEE Communications Surveys ＆ Tutorials, 2017(21): 260-288.
[42]	王浩翔 . 基于博弈论的边缘计算资源分配算法研究[D]. 邯郸:河北工程大学, 2019.
	WANG H X . Research on edge computing resource allocation algorithm based on game theory[D]. Han’dan:Hebei University of Engineering, 2019.
[43]	XU X , LIU K , DAI P ,et al. Joint task offloading and resource optimization in NOMA-based vehicular edge computing:a game-theoretic DRL approach[J]. Journal of Systems Architecture, 2023,134(1): 102780-102825.
[44]	HE Q , CUI G , ZHANG X ,et al. A game-theoretical approach for user allocation in edge computing environment[J]. IEEE Transactions on Parallel and Distributed Systems, 2019,31(3): 515-529.
[45]	CAO H , CAI J . Distributed multiuser computation offloading for cloudlet-based mobile cloud computing:a game-theoretic machine learning approach[J]. IEEE Transactions on Vehicular Technology, 2018,67(1): 752-764.
[46]	刘雷, 陈晨, 冯杰 ,等. 车载边缘计算卸载技术研究综述[J]. 电子学报, 2021,49(5): 861-871.
	LIU L , CHEN C , FENG J ,et al. Overview of research on unloading technology of on-board edge computing[J]. Journal of Electronics, 2021,49(5): 861-871.
[47]	CHEN C , JIN Y , PEI Q ,et al. A connectivity-aware intersection-based routing in VANETs[J]. EURASIP Journal on Wireless Communications and Networking, 2014,42(1): 1-16.
[48]	WAHAB O A , OTROK H , MOURAD A . VANET QoS-OLSR:QoS-based clustering protocol for vehicular Ad Hoc networks[J]. Computer Communications, 2013,36(13): 1422-1435.
[49]	WANG W , ZHAO Y , WU Q ,et al. Asynchronous federated learning based mobility-aware caching in vehicular edge computing[J]. 2022:arXiv preprint arXiv:2208.01236.
[50]	YE D , YU R , PAN M ,et al. Federated learning in vehicular edge computing:a selective model aggregation approach[J]. IEEE Access, 2020,8(1): 23920-23935.
[51]	WANG S , TUOR T , SALONIDIS T ,et al. Adaptive federated learning in resource constrained edge computing systems[J]. IEEE Journal on Selected Areas in Communications, 2019,37(6): 1205-1221.
[52]	ZENG F , REN Y , DENG X ,et al. Cost-effective edge server placement in wireless metropolitan area networks[J]. Sensors, 2018,19(1): 1-32.