Title:
ACQUISITION METHOD FOR PILOT FREQUENCY POSITION DETERMINATION MODEL, DETERMINATION METHOD FOR PILOT FREQUENCY POSITION, AND APPARATUSES THEREOF
Document Type and Number:
WIPO Patent Application WO/2023/201668
Kind Code:
A1
Abstract:
The present application may be applied to the technical field of communications, and discloses an acquisition method for a pilot frequency position determination model. The method executed by a communication device comprises: obtaining a training sample set, wherein the training sample set comprises a plurality of sample groups; and inputting the sample groups into an initial pilot frequency position determination model to output first pilot frequency position information (22); obtaining prediction channel state information on the basis of the first pilot frequency position information (23); adjusting model parameters of the initial pilot frequency position determination model on the basis of label channel state information and the prediction channel state information, and continuing to train the adjusted initial pilot frequency position determination model by using the next sample group until the training process is complete to obtain a target pilot frequency position determination model (24). By means of training the initial pilot frequency position determination model, the efficiency with which channel state information of pilot frequency array elements on an RIS is determined may be enhanced; particularly, a large number of pilot frequency array elements are present on the RIS, and therefore, the time required for the communication device to perform channel estimation may be reduced, and the accuracy may be enhanced.
Inventors:
CHI LIANGANG (CN)
LIU HAOXIANG (CN)
LIU HAOXIANG (CN)
Application Number:
PCT/CN2022/088321
Publication Date:
October 26, 2023
Filing Date:
April 21, 2022
Export Citation:
Assignee:
BEIJING XIAOMI MOBILE SOFTWARE CO LTD (CN)
International Classes:
H04L25/02; H04L27/26
Foreign References:
| CN111131096A | 2020-05-08 | |||
| CN114124173A | 2022-03-01 | |||
| EP3962006A1 | 2022-03-02 |
Other References:
MEHRAN S. ET AL.: "Pilot Pattern Design for Deep Learning-Based Channel Estimation in OFDM Systems", IEEE WIRELESS COMMUNICATIONS LETTERS, vol. 9, no. 12, 13 August 2020 (2020-08-13), XP011823718, DOI: 10.1109/LWC.2020.3016603
QUALCOMM INCORPORATED: "CSI enhancements: MTRP and FR1 FDD reciprocity", 3GPP DRAFT; R1-2103156, 3RD GENERATION PARTNERSHIP PROJECT (3GPP), MOBILE COMPETENCE CENTRE ; 650, ROUTE DES LUCIOLES ; F-06921 SOPHIA-ANTIPOLIS CEDEX ; FRANCE, vol. RAN WG1, no. e-Meeting; 20210412 - 20210420, 7 April 2021 (2021-04-07), Mobile Competence Centre ; 650, route des Lucioles ; F-06921 Sophia-Antipolis Cedex ; France, XP052177956
HEWANG WENG: "An Efficient Pilot Design of Sparse Channel Estimation for OFDM System ", JOURNAL OF HUNAN CITY UNIVERSITY(NATURAL SCIENCE), vol. 26, no. 5, 15 September 2017 (2017-09-15), pages 56 - 60, XP093102651, ISSN: 1672-7304, DOI: 10.3969/j.issn.1672-7304.2017.05.0012
QUALCOMM INCORPORATED: "CSI enhancements: MTRP and FR1 FDD reciprocity", 3GPP DRAFT; R1-2103156, 3RD GENERATION PARTNERSHIP PROJECT (3GPP), MOBILE COMPETENCE CENTRE ; 650, ROUTE DES LUCIOLES ; F-06921 SOPHIA-ANTIPOLIS CEDEX ; FRANCE, vol. RAN WG1, no. e-Meeting; 20210412 - 20210420, 7 April 2021 (2021-04-07), Mobile Competence Centre ; 650, route des Lucioles ; F-06921 Sophia-Antipolis Cedex ; France, XP052177956
HEWANG WENG: "An Efficient Pilot Design of Sparse Channel Estimation for OFDM System ", JOURNAL OF HUNAN CITY UNIVERSITY(NATURAL SCIENCE), vol. 26, no. 5, 15 September 2017 (2017-09-15), pages 56 - 60, XP093102651, ISSN: 1672-7304, DOI: 10.3969/j.issn.1672-7304.2017.05.0012
Attorney, Agent or Firm:
FASHENG INTELLECTUAL PROPERTY COMPANY, LTD. (CN)
Download PDF: