Title:
NETWORK OPTIMIZATION METHOD, DEVICE, AND STORAGE MEDIUM
Document Type and Number:
WIPO Patent Application WO/2021/000757
Kind Code:
A1
Abstract:
A network optimization method, a device, and a storage medium. The method comprises: building a model of existing problems of cells in a first area, and obtaining N intelligent agents, a model building method, and a training method, wherein a proportion of cells having the same type of problem in the cells comprised in the first area is greater than or equal to a preset threshold, the included cells have contiguous geographic locations, and an outline of the first area has a convex shape; and for each of the intelligent agents, determining an initial model of the intelligent agent according to the model building method and the training method, and using a reinforcement learning method to train the initial model of the intelligent agent; alternatively, performing network optimization on cells corresponding to the intelligent agents according to trained models of the intelligent agents; alternatively, determining initial models of the intelligent agents according to the model building method and the training method, using the reinforcement learning method to train the initial models of the intelligent agents according to the model building method and the training method, and performing network optimization on the cells corresponding to the intelligent agents.
Inventors:
XIE QINZHENG (CN)
Application Number:
PCT/CN2020/097554
Publication Date:
January 07, 2021
Filing Date:
June 22, 2020
Export Citation:
Assignee:
ZTE CORP (CN)
International Classes:
H04W16/18
Domestic Patent References:
| WO2018149898A2 | 2018-08-23 | |||
| WO2019007388A1 | 2019-01-10 |
Foreign References:
| CN106507398A | 2017-03-15 | |||
| CN101873604A | 2010-10-27 | |||
| CN106714239A | 2017-05-24 |
Other References:
See also references of EP 3979685A4
Attorney, Agent or Firm:
JIAQUAN IP LAW (CN)
Download PDF: