Title:
MACHINE LEARNING-BASED MASSIVE MIMO PROCESSING METHOD FOR MOBILE BASE STATION
Document Type and Number:
WIPO Patent Application WO/2024/090766
Kind Code:
A1
Abstract:
The present invention relates to a machine learning-based massive MIMO processing technique that achieves low power and good operational performance by adopting a scenario-based operation for a mobile base station with limitations in energy capacity and operating time. The present invention has the advantage of enhancing operational performance and operating time for a mobile base station utilizing a massive MIMO technique. In addition, the present invention has the advantage of providing good initial service even during the training of the mobile base station using machine learning. Furthermore, the present invention has the advantage of being capable of continuously performing low power neural network learning for optimizing massive MIMO processing according to changes in the operating environment of the mobile base station.
Inventors:
LEE SANG JO (KR)
Application Number:
PCT/KR2023/012760
Publication Date:
May 02, 2024
Filing Date:
August 29, 2023
Export Citation:
Assignee:
CROSSWORKS CO LTD (KR)
HFR INC (KR)
HFR INC (KR)
International Classes:
H04B7/0452; G06N20/00; H04B7/0456; H04B7/06; H04L1/20; H04W24/02
Foreign References:
| KR20220031624A | 2022-03-11 | |||
| KR20220042291A | 2022-04-05 | |||
| KR102206549B1 | 2021-01-22 | |||
| KR102355383B1 | 2022-02-08 |
Other References:
QIANG SUN: "Deep Learning-Based Joint CSI Feedback and Hybrid Precoding in FDD mmWave Massive MIMO Systems", ENTROPY, MOLECULAR DIVERSITY PRESERVATION INTERNATIONAL, BASEL, CH, vol. 24, no. 4, 23 March 2022 (2022-03-23), CH , pages 441, XP093164991, ISSN: 1099-4300, DOI: 10.3390/e24040441
Attorney, Agent or Firm:
KIM, Do Hyoung (KR)
Download PDF: