arrow
返回

Collaborative Edge AI Inference Over Cloud-RAN

delete2024-09-01
delete2
delete
OA
AI
P
Pengfei Zhang
D
Dingzhu Wen *
G
Guangxu Zhu
Q
Qimei Chen
K
Kaifeng Han
Y
Yuanming Shi
DOI:10.1109/TCOMM.2024.3388488delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
In this paper, a cloud radio access network (Cloud-RAN) based collaborative edge AI inference architecture is proposed. Specifically, geographically distributed devices capture real-time noise-corrupted sensory data samples and extract the noisy local feature vectors, which are then aggregated at each remote radio head (RRH) to suppress sensing noise. To realize efficient uplink feature aggregation, we allow each RRH receives local feature vectors from all devices over the same resource blocks simultaneously by leveraging an over-the-air computation (AirComp) technique. Thereafter, these aggregated feature vectors are quantized and transmitted to a central processor (CP) for further aggregation and downstream inference tasks. Our aim in this work is to maximize the inference accuracy via a surrogate accuracy metric called discriminant gain, which measures the discernibility of different classes in the feature space. The key challenges lie on simultaneously suppressing the coupled sensing noise, AirComp distortion caused by hostile wireless channels, and the quantization error resulting from the limited capacity of fronthaul links. To address these challenges, this work proposes a joint transmit precoding, receive beamforming, and quantization error control scheme to enhance the inference accuracy. Extensive numerical experiments demonstrate the effectiveness and superiority of our proposed optimization algorithm compared to various baselines.
Keyword:
Task analysis
Feature extraction
Collaboration
Distortion
Sensors
Computer architecture
Wireless networks
Cloud radio access network
edge AI
edge inference
over-the-air computation

期刊

IEEE Transactions on Communications 封面图
IEEE Transactions on Communications
IF:
8.3
论文数:
1.2W
被引数:
3.6W

机构

S
Shenzhen Research Institute of Big Data
学者数:
256
论文数: 352
被引数: 357
C
china academy of information & communication technology
学者数:
332
论文数: 271
被引数: 0
S
ShanghaiTech University
学者数:
9.7K
论文数: 5.9K
被引数: 1.6W
W
wuhan university
学者数:
8.1W
论文数: 5.8W
被引数: 70
学者 查看更多机构
引用论文

引用论文

Choosing Where to Attend and the Medial Frontal Cortex: An fMRI Study
err2008-09-01
err0
errOAAI
errPaul C. J. Taylor; Matthew F. S. Rushworth; Anna C. Nobre
err分享
err收藏
FRONTHAUL-CONSTRAINED CLOUD RADIO ACCESS NETWORKS: INSIGHTS AND CHALLENGES
err2015-04-01
err337
errOAAI
errPeng, Mugen; Wang, Chonggang; Lau, Vincent; Poor, H. Vincent
err分享
err收藏
Inhibitory Effects of Antioxidants on N‐Bis(2‐hydroxypropyl)nitrosamine‐induced Lung Carcinogenesis in Rats
err2005-08-25
err0
errOAAI
errRyohei Hasegawa; Fumio Furukawa; Kazuhiro Toyoda; Michihito Takahashi; Yuzo Hayashi; Masao Hirose; Nobuyuki Ito
err分享
err收藏
Finite Mixture Models有限混合模型
err2019-03-07
err559
PREAI
errMcLachlan, Geoffrey J.; Lee, Sharon X.; Rathnayake, Suren I.
err分享
err收藏
Communication-Efficient Edge AI: Algorithms and Systems通信高效的边缘AI: 算法和系统
err2020-01-01
err238
errOAAI
errShi, Yuanming; Yang, Kai; Jiang, Tao; Zhang, Jun; Letaief, Khaled B.
err分享
err收藏
学者 查看更多内容