面向联合学习的D2D计算任务卸载

doi:10.11959/j.issn.2096-3750.2019.00135

物联网学报 ›› 2019, Vol. 3 ›› Issue (4): 82-90.doi: 10.11959/j.issn.2096-3750.2019.00135

面向联合学习的D2D计算任务卸载

蔡晓然,莫小鹏,许杰

广东工业大学信息工程学院，广东广州 510006

修回日期:2019-08-09 出版日期:2019-12-30 发布日期:2020-02-05
作者简介:蔡晓然（1996- ），女，广东阳江人，广东工业大学硕士生，主要研究方向为无线通信、机器学习和移动边缘计算|莫小鹏（1996- ），男，广东湛江人，广东工业大学硕士生，主要研究方向为无线通信、机器学习、无人机通信和移动边缘计算|许杰（1985- ），男，四川内江人，广东工业大学教授、博士生导师，主要研究方向为无线通信、机器学习、无线能量传输、无人机通信以及移动边缘计算
基金资助:
国家重点研发计划资助项目(2018YFB1800800);广东省重点领域研发计划资助项目(2018B030338001)

D2D computation task offloading for efficient federated learning

Xiaoran CAI,Xiaopeng MO,Jie XU

School of Information Engineering,Guangdong University of Technology,Guangzhou 510006,China

Revised:2019-08-09 Online:2019-12-30 Published:2020-02-05
Supported by:
The National Key R＆D Program Subsidized Projects(2018YFB1800800);Research and Development Program Subsidized Projects in Key Areas of Guangdong Province(2018B030338001)

摘要/Abstract

摘要：

联合学习是一种分布式机器学习，边缘节点的计算和通信资源受限等因素是限制其性能优化的瓶颈。当边缘节点的计算和通信能力异构时，需要对通信和计算进行联合优化。提出了一种面向联合学习的D2D计算任务卸载方案，不同边缘节点通过D2D通信交换数据样本，平衡节点的处理能力和任务负载，以最小化联合学习模型训练过程的总时延。仿真结果表明，所提出的D2D计算任务卸载方案能显著提高联合学习的模型训练速度和效率。

关键词: 联合学习, 移动边缘计算, 任务卸载, D2D通信

Abstract:

Federated learning is a kind of distributed machine learning technique.The factor of communication and computation resource constraints at the edge node is becoming the performance bottleneck.In particular,when different edge node has distinct computation and communication capabilities,the model training performance may degrade severely,thus necessitating the joint communication and computation optimization.To tackle this challenge,a computational task offloading scheme enabled by device-to-device (D2D) communications was proposed,in which different edge node exchanged data samples via D2D communication links to balance the processing capability and task load,in order to minimize the total time delay for machine learning model training.Simulation results show that compared to the benchmark scheme without such D2D task offloading the training speed and efficiency of federated learning has be improved significantly.

Key words: federated learning, mobile edge computing, task offloading, device-to-device communication

中图分类号:

TN929

蔡晓然,莫小鹏,许杰. 面向联合学习的D2D计算任务卸载[J]. 物联网学报, 2019, 3(4): 82-90.

Xiaoran CAI,Xiaopeng MO,Jie XU. D2D computation task offloading for efficient federated learning[J]. Chinese Journal on Internet of Things, 2019, 3(4): 82-90.

图/表 7

图1

图2

图3

图4

图5

图6

图7

参考文献 17

[1]	MAO Y Y , YOU C S , ZHANG J ,et al. A survey on mobile edge computing:the communication perspective[J]. IEEE Communications Surveys ＆ Tutorials, 2017,19(4): 2322-2358.
[2]	CAO X W , WANG F , XU J ,et al. Joint computation and communication cooperation for energy-efficient mobile edge computing[J]. IEEE Internet of Things Journal, 2019,6(3): 4188-4200.
[3]	WANG F , XU J , WANG X ,et al. Joint offloading and computing optimization in wireless powered mobile-edge computing systems[J]. IEEE Transactions on Wireless Communications, 2018,17(3): 1784-1797.
[4]	LETAIEF B K , CHEN W , SHI Y M ,et al. The roadmap to 6G:AI empowered wireless networks[J]. IEEE Communications Magazine, 2019,57(8): 84-90.
[5]	ELGAZAR A , HARRAS K , AAZAM M ,et al. Towards intelligent edge storage management:determining and predicting mobile file popularity[C]// IEEE International Conference on Mobile Cloud Computing,Services,and Engineering. IEEE, 2018.
[6]	KONECNY J , BRENDAN M , RAMAGE D ,et al. Federated machine learning:concept and applications[J]. ACM Transactions on Intelligent Systems and Technology, 2019,10(2):12.
[7]	ZHANG J L , TU H D , REN Y J ,et al. An adaptive synchronous parallel strategy for distributed machine learning[J]. IEEE Access, 2018(6): 19222-19230.
[8]	SPRAGUE M R , JALALIRAD A , SCAVUZZO M ,et al. Asynchronous federated learning for geospatial applications[C]// Joint European Conference on Machine Learning and Knowledge Discovery in Databases. Springer, 2018.
[9]	CHEN J M , MONGA R , BENGIO S ,et al. Revisiting distributed synchronous SGD[C]// International Conference on Learning Representations Workshop Track. Arxiv, 2016.
[10]	WANG S Q , 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.
[11]	WEN W , XU C , YAN F ,et al. TernGrad:ternary gradients to reduce communication in distributed deep learning[C]// Advances in Neural Information Processing Systems. IEEE, 2017.
[12]	LI X Y , ZHU G X , GONG Y ,et al. Wirelessly powered data aggregation for IoT via over-the-air function computation:beamforming and power control[J]. IEEE Transactions on Wireless Communications, 2019,18(7): 3437-3452.
[13]	HU Y C , PATEL M , SABELLA D ,et al. Mobile edge computing—a key technology towards 5G[J]. ETSI White Paper, 2015,11(11): 1-16.
[14]	XU J , CHEN L X , LIU K ,et al. Designing security-aware incentives for computation offloading via device-to-device communication[J]. IEEE Transactions on Wireless Communications, 2018,17(9): 6053-6066.
[15]	LEE S J , TCHA Y , SEO S Y ,et al. Efficient use of multicast and unicast channels for multicast service transmission[J]. IEEE Transactions on Communications, 2011,59(5): 1264-1267.
[16]	HUNGER R . Floating point operations in matrix-vector calculus[M]. Munich: Institute for Circuit Theory and Signal ProcessingPress, 2005.
[17]	LECUN Y , BOTTOU L , BENGIO Y ,et al. Gradient-based learning applied to document recognition[J]. IEEE, 1998,86(11): 2278-2324.

面向联合学习的D2D计算任务卸载

D2D computation task offloading for efficient 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 7

参考文献 17

相关文章 15

Metrics

推荐阅读 0

[1]	刘耀, 何岳园, 周红静, 李超良, 李闯. 移动边缘计算中基于资源联合分配的部分计算卸载方法[J]. 物联网学报, 2023, 7(1): 140-148.
[2]	苏麟, 党小超, 郝占军, 汝春瑞, 尚旭. 基于WPT-MEC的动态自适应卸载方法[J]. 物联网学报, 2022, 6(4): 128-138.
[3]	李贤, 毕宿志, 曾泓儒, 林彬, 林晓辉. 基于智能化用户协作的边缘计算任务卸载与资源分配优化[J]. 物联网学报, 2022, 6(4): 41-52.
[4]	方娟, 叶志远, 张梦媛, 史佳眉, 滕自怡. 边云协同场景下基于强化学习的精英分层任务卸载策略研究[J]. 物联网学报, 2022, 6(1): 91-100.
[5]	苏新, 江苏, 周一青. 面向海洋观监测传感网的移动终端位置隐私保护研究[J]. 物联网学报, 2021, 5(4): 26-36.
[6]	路静, 李晗琳, 高林. 移动边缘计算任务切分与最优卸载算法设计[J]. 物联网学报, 2021, 5(2): 78-86.
[7]	张厚浩, 李晗琳, 高林. 移动边缘计算中的分层资源部署与共享策略[J]. 物联网学报, 2021, 5(1): 11-18.
[8]	嵇介曲, 朱琨, 易畅言, 王然. 多无人机辅助移动边缘计算中的任务卸载和轨迹优化[J]. 物联网学报, 2021, 5(1): 27-35.
[9]	周鹏,徐金城,杨博. 工业物联网中基于边缘计算的跨域计算资源分配与任务卸载[J]. 物联网学报, 2020, 4(2): 96-104.
[10]	王鹏飞,邸博雅,宋令阳,韩竹. 6G异构边缘计算[J]. 物联网学报, 2020, 4(1): 121-130.
[11]	宋富洪,邢焕来,潘炜. 一种面向移动云计算的多目标任务卸载算法[J]. 物联网学报, 2019, 3(3): 41-49.
[12]	曾启程,孙宇璇,周盛. 车辆间计算任务卸载算法与系统级仿真验证[J]. 物联网学报, 2019, 3(3): 62-69.
[13]	乔冠华,冷甦鹏,刘浩,黄开胜,吴凡. 面向车辆多址接入边缘计算网络的任务协同计算迁移策略[J]. 物联网学报, 2019, 3(1): 51-59.
[14]	王璐瑶,张文倩,张光林. 多用户移动边缘计算迁移的能量管理研究[J]. 物联网学报, 2019, 3(1): 73-81.
[15]	丛书畅,姚超,王鹏飞,郑子杰,宋令阳. EdgeFlow移动边缘计算在物联网中的应用研究[J]. 物联网学报, 2019, 3(1): 90-96.