A RIS-assisted multi-user communication beamforming optimization method and system based on graph neural network

By optimizing the beamforming of the base station and RIS using graph neural networks, the problems of imperfect channel state information and high complexity in RIS-assisted multi-user communication are solved, thereby maximizing the downlink transmission rate and improving the communication performance of the system.

CN120049928BActive Publication Date: 2025-10-28SOUTHEAST UNIV
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202510187972.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-02-20
Publication Date
2025-10-28
Estimated Expiration
2045-02-20

AI Technical Summary

Technical Problem

Existing technologies in RIS-assisted multi-user multiple-input single-output systems fail to effectively consider imperfect channel state information, and traditional iterative optimization algorithms are highly complex, lacking low-complexity joint beamforming optimization methods.

Method used

A graph neural network-based approach is adopted to optimize the active beamforming of the base station and the passive beamforming of the RIS, and to utilize the reciprocity of the uplink and downlink channels to construct a graph neural network model, thereby jointly optimizing the beamforming of the base station and the RIS to maximize the downlink transmission rate of the system.

Benefits of technology

Without relying on perfect channel state information, the downlink transmission rate of the RIS-assisted MU-MISO system was maximized, reducing optimization complexity and improving communication performance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120049928B_ABST
    Figure CN120049928B_ABST
Patent Text Reader

Abstract

This invention discloses a RIS-assisted multi-user communication beamforming optimization method and system based on graph neural networks, belonging to the field of reconfigurable intelligent metasurface-assisted communication. This invention considers the scenario of RIS-assisted multi-user communication, utilizing graph neural networks to optimize the active beamforming of the base station and the passive beamforming of the RIS, thereby maximizing the downlink system transmission rate. The graph neural network model includes one RIS node and user nodes with the same number of serving users. The RIS node is responsible for obtaining the passive beamforming for the RIS using graph learning based on the channel state estimation information input from all user nodes. Each user node corresponds to one user, and the active beamforming of the base station corresponding to that user is obtained using graph learning based on the channel state estimation information input from that user. This invention uses graph neural networks to effectively solve the joint optimization problem of active and passive beamforming without requiring perfect channel estimation.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Secure communication optimization method for double-IRS-assisted simultaneous wireless information and power transfer system based on graph neural network

    CN119300018A