Communication-Efficient Edge AI Inference over Wireless Networks

doi:10.12142/ZTECOM.202002005

ZTE Communications ›› 2020, Vol. 18 ›› Issue (2): 31-39.DOI: 10.12142/ZTECOM.202002005

• Special Topic • Previous Articles Next Articles

Communication-Efficient Edge AI Inference over Wireless Networks

YANG Kai, ZHOU Yong(), YANG Zhanpeng, SHI Yuanming

School of Information Science and Technology, ShanghaiTech University, Shanghai 201210, China

Received:2020-02-10 Online:2020-06-25 Published:2020-08-07
About author:YANG Kai received the B.S. degree in electronic engineering from Dalian University of Technology, China in 2015. He is currently working toward the Ph.D. degree with the School of Information Science and Technology, ShanghaiTech University, China, also with the Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, China, and also with the University of Chinese Academy of Sciences, Beijing, China. His research interests include data processing and optimization for mobile edge artificial intelligence.|ZHOU Yong (zhouyong@shanghaitech.edu.cn) received the B.S. and M.Eng. degrees from Shandong University, China in 2008 and 2011, respectively, and the Ph.D. degree from The University of Waterloo, Canada in 2015. From November 2015 to January 2018, he worked as a postdoctoral research fellow in the Department of Electrical and Computer Engineering, The University of British Columbia, Canada. Since March 2018, he has been with the School of Information Science and Technology, ShanghaiTech University, where he is currently an assistant professor. His research interests include 5G and beyond, IoT, and edge AI.|YANG Zhanpeng will receive his B.S. degree from Xidian University, China on July 2020. He will join the School of Information Science and Technology, ShanghaiTech University, in Fall 2020. He mainly focuses on developed reconfigurable intelligence surface based 6G wireless technologies for mobile edge AI systems.|SHI Yuanming received the B.S. degree in electronic engineering from Tsinghua University, China in 2011. He received the Ph.D. degree in electronic and computer engineering from The Hong Kong University of Science and Technology (HKUST), China in 2015. Since September 2015, he has been with the School of Information Science and Technology, ShanghaiTech University, China, where he is currently a tenured associate professor. He visited University of California, Berkeley, USA from October 2016 to February 2017. Dr. SHI is a recipient of the 2016 IEEE Marconi Prize Paper Award in Wireless Communications and the 2016 Young Author Best Paper Award by the IEEE Signal Processing Society. He is an editor of IEEE Transactions on Wireless Communications. His research areas include optimization, statistics, machine learning, signal processing, and their applications to 6G, IoT, AI and FinTech.

Abstract

Abstract:

Given the fast growth of intelligent devices, it is expected that a large number of high-stakes artificial intelligence (AI) applications, e.g., drones, autonomous cars, and tactile robots, will be deployed at the edge of wireless networks in the near future. Therefore, the intelligent communication networks will be designed to leverage advanced wireless techniques and edge computing technologies to support AI-enabled applications at various end devices with limited communication, computation, hardware and energy resources. In this article, we present the principles of efficient deployment of model inference at network edge to provide low-latency and energy-efficient AI services. This includes the wireless distributed computing framework for low-latency device distributed model inference as well as the wireless cooperative transmission strategy for energy-efficient edge cooperative model inference. The communication efficiency of edge inference systems is further improved by building up a smart radio propagation environment via intelligent reflecting surface.

Key words: communication efficiency, cooperative transmission, distributed computing, edge AI, edge inference

YANG Kai, ZHOU Yong, YANG Zhanpeng, SHI Yuanming. Communication-Efficient Edge AI Inference over Wireless Networks[J]. ZTE Communications, 2020, 18(2): 31-39.

Figures/Tables 7

Figure 1 Illustration of edge AI.

Figure 2 Illustration of computing model of MapReduce-based distributed computing framework.

Figure 3 Illustration of communication model for data shuffling of on-device distributed AI inference systems.

Figure 4 Achievable DoF of algorithms versus the number of devices for the data shuffling of on-device distributed AI inference systems.

Figure 5 Illustration of in-edge cooperative AI inference systems.

Figure 6 Average total power consumption comparison between edge processing systems with and without IRS.

Figure 7 Average total power consumption comparison for different algorithms in edge processing systems.

References 22

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 (Fourthquarter): 2322–2358. DOI: 10.1109/COMST.2017.2745201 DOI
2	LETAIEF K B, CHEN W, SHI Y M, et al. The roadmap to 6G: AI empowered wireless networks [J]. IEEE communications magazine, 2019, 57(8): 84–90. DOI: 10.1109/MCOM.2019.1900271 DOI
3	YANG K, SHI Y M, DING Z. Data shuffling in wireless distributed computing via low⁃rank optimization [J]. IEEE transactions on signal processing, 2019, 67(12): 3087–3099. DOI:10.1109/tsp.2019.2912139 DOI
4	LI S Z, MADDAH⁃ALI M A, YU Q, et al. A fundamental tradeoff between computation and communication in distributed computing [J]. IEEE transactions on information theory, 2018, 64(1): 109–128. DOI: 10.1109/TIT.2017.2756959 DOI
5	LI S Z, MADDAH⁃ALI M A, AVESTIMEHR A S. Coding for distributed fog computing [J]. IEEE communications magazine, 2017, 55(4): 34–40. DOI: 10.1109/mcom.2017.1600894 DOI
6	GESBERT D, HANLY S, HUANG H, et al. Multi⁃cell MIMO cooperative networks: a new look at interference [J]. IEEE journal on selected areas in communications, 2010, 28(9): 1380–1408. DOI: 10.1109/jsac.2010.101202 DOI
7	YUAN X J, ZHANG Y⁃J, SHI Y M, et al. Reconfigurable⁃intelligent⁃surface empowered 6G wireless communications: challenges and opportunities [EB/OL]. (2020⁃01⁃02).
8	ZHOU Z, CHEN X, LI E, et al. Edge intelligence: paving the last mile of artificial intelligence with edge computing [J]. Proceedings of the IEEE, 2019, 107(8): 1738–1762. DOI: 10.1109/jproc.2019.2918951 DOI
9	PARK J, SAMARAKOON S, BENNIS M, et al. Wireless network intelligence at the edge [J]. Proceedings of the IEEE, 2019, 107(11): 2204–2239. DOI: 10.1109/jproc.2019.2941458 DOI
10	YANG K., SHI Y M, YU W, et al. Energy⁃efficient processing and robust wireless cooperative transmission for edge inference [J]. IEEE internet of things journal, 2020. DOI: 10.1109/JIOT.2020.2979523 DOI
11	HUA S, ZHOU Y, YANG K, et al. Reconfigurable intelligent surface for green edge inference [EB/OL]. (2019⁃12⁃02).
12	LI E, ZENG L K, ZHOU Z, et al. Edge AI: On⁃demand accelerating deep neural network inference via edge computing [J]. IEEE transactions on wireless communications, 2020, 19(1): 447–457. DOI: 10.1109/twc.2019.2946140 DOI
13	ESHRATIFAR A E, ABRISHAMI M S, PEDRAM M. JointDNN: an efficient training and inference engine for intelligent mobile cloud computing services [J]. IEEE transactions on mobile computing, 2019: 1. DOI: 10.1109/tmc.2019.2947893 DOI
14	ZHANG J, LETAIEF K B. Mobile edge intelligence and computing for the internet of vehicles [J]. Proceedings of the IEEE, 2020, 108(2): 246–261. DOI: 10.1109/jproc.2019.2947490 DOI
15	ZENG Y, WU Q Q, ZHANG R. Accessing from the sky: a tutorial on UAV communications for 5g and beyond [J]. Proceedings of the IEEE, 2019, 107(12): 2327–2375. DOI: 10.1109/jproc.2019.2952892 DOI
16	HADDADIN S, JOHANNSMEIER L, LEDEZMA F D. Tactile robots as a central embodiment of the tactile Internet [J]. Proceedings of the IEEE, 2019, 107(2): 471–487. DOI: 10.1109/JPROC.2018.2879870 DOI
17	LIU K Q, TAO M X. Generalized signal alignment: on the achievable DoF for multi⁃user MIMO two⁃way relay channels [J]. IEEE transactions on information theory, 2015, 61(6): 3365–3386. DOI: 10.1109/tit.2015.2420100 DOI
18	TAO P D, AN L T H. Convex analysis approach to DC programming: theory, algorithms and applications [J]. Acta mathematica vietnamica, 1997, 22(1): 289–355
19	LECUN Y, BENGIO Y, HINTON G. Deep learning [J]. Nature, 2015, 521(7553): 436–444. DOI: 10.1038/nature14539 DOI
20	LI K K, TAO M X, CHEN Z Y. Exploiting computation replication for mobile edge computing: a fundamental computation⁃communication tradeoff study [EB/OL]. (2019⁃03⁃26).
21	YANG T⁃J, CHEN Y⁃H, SZE V. Designing energy⁃efficient convolutional neural networks using energy⁃aware pruning [C]//IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu, USA, 2017: 5687–5695. DOI: 10.1109/cvpr.2017.643 DOI
22	GONG S M, LU X, HOANG D T, et al. Towards smart radio environment for wireless communications via intelligent reflecting surfaces: a comprehensive survey [EB/OL]. (2019⁃12⁃17).

[1]	YANG Howard H., ZHAO Zhongyuan, QUEK Tony Q. S.. Enabling Intelligence at Network Edge:An Overview of Federated Learning [J]. ZTE Communications, 2020, 18(2): 2-10.
[2]	SHI Lei, ZHAO Liang, SONG Wenzhan, Goutham Kamath, WU Yuan, LIU Xuefeng. Distributed Least-Squares Iterative Methods in Large-Scale Networks: A Survey [J]. ZTE Communications, 2017, 15(3): 37-45.
[3]	Zhao Pei, Lu Ping, Luo Shengmei. Cloud Computing Technology and Its Applications [J]. ZTE Communications, 2010, 8(4): 34-38.

Communication-Efficient Edge AI Inference over Wireless Networks

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 7

References 22

Related Articles 3

Recommended Articles

Metrics