GNN-Based Data Rate Maximization in Double Intelligent Reflecting Surface (IRS)-Aided Communication System

Kaixin Li, Limei Peng, Pin Han Ho

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

Abstract

This paper explores the utilization of double intelligent reflecting surfaces (IRSs) in wireless communication systems to enhance signal propagation. By dynamically adjusting the reflecting elements of the IRSs, we can efficiently manipulate the signal direction to improve the communication performance such as data rate. Nonetheless, due to the passive nature of the reflecting elements, it is quite challenging to accurately measure the channels directly between the base station (BS), the IRS, and the users based on the conventional channel estimation methods. To address this, we propose to apply a machine learning algorithm that bypasses the traditional channel estimation process and optimizes system parameters by directly extracting information from the received pilot signal. Through extensive simulations, we demonstrate the effectiveness of our proposed method in enhancing the performance of wireless systems incorporating double IRSs.

Original languageEnglish
Title of host publicationProceedings - 2023 International Conference on Networking and Network Applications, NaNA 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages447-452
Number of pages6
ISBN (Electronic)9798350327380
DOIs
StatePublished - 2023
Event2023 International Conference on Networking and Network Applications, NaNA 2023 - Qingdao, China
Duration: 18 Aug 202321 Aug 2023

Publication series

NameProceedings - 2023 International Conference on Networking and Network Applications, NaNA 2023

Conference

Conference2023 International Conference on Networking and Network Applications, NaNA 2023
Country/TerritoryChina
CityQingdao
Period18/08/2321/08/23

Keywords

  • graph neural network (GNN)
  • Intelligent reflecting surface (IRS)
  • reconfig-urable intelligent surfaces (RIS)

Fingerprint

Dive into the research topics of 'GNN-Based Data Rate Maximization in Double Intelligent Reflecting Surface (IRS)-Aided Communication System'. Together they form a unique fingerprint.

Cite this