NWDAF Model Training Separation for Flexible Network Data Analysis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The flexibility of the model training function in the Network Data Analytics Function (NWDAF) entity is not sufficient.

Innovation Solution

The NWDAF entity is designed to request and receive models from separate training platforms, modules, or service modules, allowing these entities to be used by multiple NWDAFs, enhancing flexibility.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the NWDAF entity includes integrated model training function, then the entity can perform complete data analysis operations, but the usage of model training function is not flexible enough

Engineering Contradiction:

Improveflexibility of model training functionVSAvoidentity structure complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent divides the NWDAF entity into separate functional modules: the inference platform and the training platform are independently deployed. The inference platform handles model inference operations while the training platform handles model training operations. This segmentation allows the model training function to be independently managed and shared across multiple NWDAF entities, thereby improving flexibility without significantly increasing overall system complexity.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The training platform is designed as a universal component that can serve multiple NWDAF entities. Instead of each NWDAF entity having its own dedicated training function, the training platform provides model training services to multiple inference platforms, enabling resource sharing and improving the adaptability and versatility of the model training function across the network.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS12520168B2Network data analysis method, functional entity and electronic device

Publication Date: 2026.01.06 CHINA MOBILE COMM GRP CO LTD

AI summary

A network data analysis method, a functional entity and an electronic device, the network data analysis method comprising: requesting that a first object generate a first model (101), the first object being a training platform, a training module, a training functional entity or a training service module; receiving a model sent by the first object (102).