Edge Learning for B5G Networks With Distributed Signal Processing: Semantic Communication, Edge Computing, and Wireless Sensing
Граничное обучение для сетей B5G с распределённой обработкой сигналов: семантическая связь, периферийные вычисления и беспроводное зондирование
2023-01-01
SCID: 54.1/ews5xzkb
Discuss with AI
distributed signal processingedge computingedge learningsemantic communicationwireless sensing
Figures from the paper
Abstract (AI)
To process and transfer large amounts of data in emerging wireless services, it has become increasingly appealing to exploit distributed data communication and learning. Specifically, edge learning (EL) enables local model training on geographically disperse edge nodes and minimizes the need for frequent data exchange. However, the current design of separating EL deployment and communication optimization does not yet reap the promised benefits of distributed signal processing, and sometimes suffers from excessive signalling overhead, long processing delay, and unstable learning convergence. In this paper, we provide an overview on practical distributed EL techniques and their interplay with advanced communication optimization designs. In particular, typical performance metrics for dual-functional learning and communication networks are discussed. Also, recent achievements of enabling techniques for the dual-functional design are surveyed with exemplifications from the mutual perspectives of “communications for learning” and “learning for communications.” The application of EL techniques within a variety of future communication systems are also envisioned for beyond 5G (B5G) wireless networks. For the application in goal-oriented semantic communication, we present a first mathematical model of the goal-oriented source entropy as an optimization problem. In addition, from the viewpoint of information theory, we identify fundamental open problems of characterizing rate regions for communication networks supporting distributed learning-and-computing tasks. We also present technical challenges as well as emerging application opportunities in this field, with the aim of inspiring future research and promoting widespread developments of EL in B5G.
Key Findings
1
For goal-oriented semantic communication, the paper presents a first mathematical formulation of goal-oriented source entropy as an optimization problem.
2
It organizes enabling approaches through the complementary perspectives of communications for learning and learning for communications, including applications in future B5G systems.
3
Separately designing edge learning deployment and communication optimization can cause excessive signaling overhead, long processing delays, and unstable learning convergence.
4
The paper identifies information-theoretic open problems in characterizing rate regions for networks supporting distributed learning and computing tasks.
5
The paper surveys practical distributed edge-learning techniques and their integration with advanced communication optimization for dual-functional learning-and-communication networks.
Research Object
Edge learning in beyond-5G (B5G) wireless networks with distributed signal processing
Research Subject
The interplay between distributed edge-learning and communication optimization, including performance, signaling overhead, processing delay, learning convergence, semantic communication, edge computing, and wireless sensing
Publication Details
Publication Date
2023-01-01
Journal
Publisher
ISSN
Cited by
776
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
References available in scid.ai6
Elements of Information Theory2001
A Survey on Mobile Edge Computing: The Communication Perspective2017
Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation2016
Edge Artificial Intelligence for 6G: Vision, Enabling Technologies, and Applications2021
End-to-End Evaluation of Federated Learning and Split Learning for Internet of Things2020
Detailed comparison of communication efficiency of split learning and federated learning2019