Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
WIRELESS POLICY OPTIMIZATION METHOD AND APPARATUS
Document Type and Number:
WIPO Patent Application WO/2024/065476
Kind Code:
A1
Abstract:
A wireless policy optimization method, comprising: acquiring a graph representation of a wireless communication network, wherein the graph representation comprises a plurality of nodes and edges for connecting the plurality of nodes, the plurality of nodes comprise a first node and a second node, and the first node and the second node are connected to each other by means of an edge; taking the graph representation as an input of a graph neural network, and outputting final feature vectors of the edges; and guiding the optimization of a wireless policy on the basis of the final feature vectors of the edges, wherein the structure of the graph neural network is constructed on the basis of guidance of Taylor expansion of a matrix inversion operation. Therefore, the problem of the traditional GNN having relatively poor generalization performance is solved.

Inventors:
GUO JIA (CN)
YANG CHENYANG (CN)
WANG JIAN (CN)
LI RONG (CN)
Application Number:
PCT/CN2022/122835
Publication Date:
April 04, 2024
Filing Date:
September 29, 2022
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
HUAWEI TECH CO LTD (CN)
International Classes:
H04L27/00; G06N3/04; G06N3/08
Domestic Patent References:
WO2022125270A12022-06-16
Foreign References:
CN114786258A2022-07-22
CN114707641A2022-07-05
CN114679355A2022-06-28
CN111465944A2020-07-28
US20220124543A12022-04-21
Other References:
JIA GUO: "A Model-based GNN for Learning Precoding", ARXIV, CORNELL UNIVERSITY LIBRARY, ARXIV.ORG, ITHACA, 1 December 2022 (2022-12-01), XP093152349, Retrieved from the Internet DOI: 10.48550/arxiv.2212.00860
Attorney, Agent or Firm:
E-TONE INTELLECTUAL PROPERTY FIRM (GENERAL PARTNERSHIP) (CN)
Download PDF: