Machine learning system for predicting resource usage of 5G network slices or cells
By training machine learning systems to cluster and standardize the time series data of 5G network slices or cells, the problem of improper resource allocation in the existing technology is solved, and efficient management and dynamic adjustment of 5G networks are achieved.
CN120238913APending Publication Date: 2025-07-01JUNIPER NETWORKS INC
View PDF 0 Cites 0 Cited by
Patent Information
- Application Number
- CN202411543182.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2023-12-29
- Filing Date
- 2024-10-31
- Publication Date
- 2025-07-01
Smart Images
Figure CN120238913A_ABST
Abstract
The invention relates to a machine learning system for predicting resource usage of 5G network slices or cells. A machine learning system is trained to predict resource usage for cells or network slices of a mobile network. For example, a computing system obtains respective data sets of cells or network slices. Each data set includes a time step and a corresponding value of a performance indicator for a corresponding one of the cell or the network slice. A computing system groups datasets into dataset clusters based on a clustering algorithm applied to (1) the datasets or (2) cells or network slices. The computing system applies a transform to a subset of recent time steps and corresponding values of each data set of a first of the clusters to obtain a set of time steps and corresponding normalized values with which the machine learning system is trained to generate predicted values at future time steps of the data set of the first cluster.
Need to check novelty before this filing date? Find Prior Art