SEMUAV: A Cooperative Semantic Communication Method in SATIN

Shaobo Ma, Wei Liang, Boxuan Zhang, Xuan Xue, Ang Gao

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

With the development of sixth-generation (6G) networks, semantic communication has been recognized as an emerging technology that could provide intelligent Communication, where message is transmitted with less data volume, alleviating the problem of resource constraints. At the same time, the shared unique knowledge base between the transmitter and receiver reduces the risk of messages being intercepted, since the interceptor cannot decode them. In this paper, we combined semantic communication with SATIN and proposed UAV-assisted communication methods named SEMUAV-AF as well as SEMUAV-DF. Meanwhile, we designed a SNR-Adaptive module and a lightweight model for better practical deployment. The simulation results prove that the proposed model is more efficient than the communication without UAV assistance, while adaptive module and lightweight model could also meet expec-tations. Overall, our proposed approach of cooperative semantic communication could improve the performance of SATIN system while ensuring information security.

Original languageEnglish
Title of host publication2023 IEEE Globecom Workshops, GC Wkshps 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages2129-2134
Number of pages6
ISBN (Electronic)9798350370218
DOIs
StatePublished - 2023
Event2023 IEEE Globecom Workshops, GC Wkshps 2023 - Kuala Lumpur, Malaysia
Duration: 4 Dec 20238 Dec 2023

Publication series

Name2023 IEEE Globecom Workshops, GC Wkshps 2023

Conference

Conference2023 IEEE Globecom Workshops, GC Wkshps 2023
Country/TerritoryMalaysia
CityKuala Lumpur
Period4/12/238/12/23

Keywords

  • Cooperative Communication
  • Deep Learning
  • SATIN
  • Semantic Communication
  • UAV

Fingerprint

Dive into the research topics of 'SEMUAV: A Cooperative Semantic Communication Method in SATIN'. Together they form a unique fingerprint.

Cite this