AI-Predicted QoS Coordination Between Access and Core Networks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The QoS parameters generated by the core network often fail to meet the service requirements on the access network side, leading to inefficiencies in service processing due to network latency and congestion.

Innovation Solution

An access network device uses an AI model to predict QoS association information, which is sent to the core network to determine a QoS parameter set that better aligns with the access network's requirements, allowing for efficient QoS parameter acquisition.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If QoS parameters are generated by the core network, then the QoS mechanism can be implemented, but the QoS parameters may not meet the service requirements on the access network side

Engineering Contradiction:
ImproveQoS parameter suitabilityVSAvoidservice processing efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The access network device performs AI model prediction in advance to generate QoS association information before the core network generates QoS parameters. This preliminary action ensures that the generated parameters are more likely to meet access network service requirements, avoiding rework and improving overall efficiency

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The access network device sends QoS association information to the core network, creating a feedback loop that allows the core network to generate QoS parameters better aligned with access network requirements. This feedback mechanism resolves the contradiction by enabling continuous improvement of parameter suitability

Inventive Principle:
Principle #23Feedback

2Reliability

If QoS parameters are delivered from core network to access network device, then the QoS mechanism can operate, but the parameters may not align with access network service requirements

Engineering Contradiction:
ImproveQoS parameter accuracyVSAvoidparameter acquisition time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The access network device prepares QoS association information using AI prediction before the core network generates parameters. This preliminary preparation reduces the time needed for parameter acquisition and iteration, while improving accuracy through informed prediction

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The access network device autonomously generates QoS association information using its own AI model based on local service requirements. This self-service capability reduces dependency on core network parameter delivery and minimizes acquisition time

Inventive Principle:
Principle #25Self-service

Data Source

PatentEP4712547A1Communication method, apparatus and system, and storage medium and program product
Publication Date: 2026.03.18 HUAWEI TECH CO LTD
  • EP4712547A1 patent drawingFigure 1
  • EP4712547A1 patent drawingFigure 2~3
  • EP4712547A1 patent drawingFigure 4

AI summary

This application discloses a communication method, apparatus, and system, a storage medium, and a program product. A core network element may obtain QoS association information predicted by an access network device, obtain a first QoS parameter set through prediction based on the QoS association information, and send the first QoS parameter set to the access network device. Because the first QoS parameter set is obtained based on the QoS association information predicted by the access network device, a service requirement on an access network side can be better met, so that the access network device can efficiently obtain an available QoS parameter, and efficiency of performing service processing by using the QoS parameter is improved.