Title:
NEURAL NETWORK-BASED POWER DISTRIBUTION METHOD AND DEVICE
Document Type and Number:
WIPO Patent Application WO/2020/191686
Kind Code:
A1
Abstract:
A neural network-based power distribution method and device. The method comprises: using a neural network model to extract communication system features on the basis of channel vector features of a MIMO communication system; further using the neural network model to perform multiple iterations of fitting for the features; and determining, according to a fitting result, a transmission power to be distributed to respective users. The invention takes into consideration features of a communication system when distributing power, thereby optimizing the result of transmission power distribution.
Inventors:
HUANG HONGJI (CN)
HU HUI (CN)
LIU JINNAN (CN)
YANG FAN (CN)
HU HUI (CN)
LIU JINNAN (CN)
YANG FAN (CN)
Application Number:
PCT/CN2019/079957
Publication Date:
October 01, 2020
Filing Date:
March 27, 2019
Export Citation:
Assignee:
HUAWEI TECH CO LTD (CN)
International Classes:
H04W72/04; G05B13/04; G06N20/00
Domestic Patent References:
WO2018193360A1 | 2018-10-25 |
Foreign References:
CN109474980A | 2019-03-15 | |||
CN109474980A | 2019-03-15 | |||
CN108462517A | 2018-08-28 | |||
CN108462517A | 2018-08-28 | |||
US8595162B2 | 2013-11-26 |
Attorney, Agent or Firm:
TDIP & PARTNERS (CN)
Download PDF:
Previous Patent: FREQUENCY ADJUSTMENT METHOD AND APPARATUS APPLIED TO TERMINAL, AND ELECTRONIC DEVICE
Next Patent: DATA PROCESSING METHOD AND DEVICE
Next Patent: DATA PROCESSING METHOD AND DEVICE