Task-Oriented Sensing, Computation, and Communication Integration for Multi-Device Edge AI

doi:10.1109/ICC45041.2023.10279277

	Task-Oriented Sensing, Computation, and Communication Integration for Multi-Device Edge AI
	Wen, Dingzhu1 ; Liu, Peixi 2,5; Zhu, Guangxu 5; Shi, Yuanming1 ; Xu, Jie 4; Eldar, Yonina C.3; Cui, Shuguang 4,5,6
	2023
会议录名称	IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS
ISSN	1550-3607
卷号	2023-May
页码	3608-3613
DOI	10.1109/ICC45041.2023.10279277
摘要	This paper studies a new multi-device edge artificial-intelligent (AI) system, which jointly exploits the AI model split inference and integrated sensing and communication (ISAC) to enable low-latency intelligent services at the network edge. In this system, multiple ISAC devices perform radar sensing to obtain multi-view data, and then offload the quantized version of extracted features to a centralized edge server, which conducts model inference based on the cascaded feature vectors. Under this setup and by considering classification tasks, we measure the inference accuracy by adopting an approximate but tractable metric, namely discriminant gain, which is defined as the distance of two classes in the Euclidean feature space under normalized covariance. To maximize the discriminant gain, we first quantify the influence of the sensing, computation, and communication processes on it with a derived closed-form expression. Then, an end-to-end task-oriented resource management approach is developed by designing an optimal integrated sensing, computation, and communication (ISCC) scheme. By using human motions recognition as a concrete AI inference task, extensive experiments are conducted to verify the performance of the proposed scheme. © 2023 IEEE.
关键词	Artificial intelligent Communication integration Integrated sensing Intelligent models Intelligent Services Low latency Multi-devices Network edges Sensing devices Task-oriented
会议名称	2023 IEEE International Conference on Communications, ICC 2023
会议地点	Rome, Italy
会议日期	May 28, 2023 - June 1, 2023
URL	查看原文
收录类别	EI
语种	英语
出版者	Institute of Electrical and Electronics Engineers Inc.
EI入藏号	20234815114293
原始文献类型	Conference article (CA)
引用统计	正在获取...
文献类型	会议论文
条目标识符	https://kms.shanghaitech.edu.cn/handle/2MSLDSTB/348757
专题	信息科学与技术学院信息科学与技术学院_PI研究组_石远明组信息科学与技术学院_PI研究组_文鼎柱组
通讯作者	Wen, Dingzhu; Shi, Yuanming
作者单位	1.School of Information Science and Technology, ShanghaiTech University, Shanghai, China 2.School of Electronics, Peking University, Beijing, China 3.Weizmann Institute of Science, Rehovot, Israel 4.School of Science and Engineering (SSE), The Future Network of Intelligence Institute (FNii), The Guangdong Provincial Key Laboratory of Future Networks of Intelligence, The Chinese University of Hong Kong, Shenzhen, China 5.Shenzhen Research Institute of Big Data, Shenzhen, China 6.Peng Cheng Laboratory, Shenzhen, China
第一作者单位	信息科学与技术学院
通讯作者单位	信息科学与技术学院
第一作者的第一单位	信息科学与技术学院
推荐引用方式 GB/T 7714	Wen, Dingzhu,Liu, Peixi,Zhu, Guangxu,et al. Task-Oriented Sensing, Computation, and Communication Integration for Multi-Device Edge AI[C]:Institute of Electrical and Electronics Engineers Inc.,2023:3608-3613.