AI Service Configuration in Communication Networks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current wireless communication systems lack integration of artificial intelligence (AI) computing power, necessitating a method to incorporate AI capabilities into communication networks.

Innovation Solution

A communication method and device that provide configuration information for AI services, enabling terminal devices to transmit service information based on these configurations, thereby integrating AI computing power into the communication network.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If AI computing power is integrated into the communication network, then the functionality and versatility of the network is improved, but the system complexity increases

Engineering Contradiction:
ImproveAI service integration capabilityVSAvoidconfiguration management complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments AI service configuration into multiple independent information elements including AI service identity, AI service type, AI model information, training data information, and hyperparameter information. This segmentation allows each aspect of AI service configuration to be managed separately, reducing overall system complexity while maintaining versatility.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements dynamic configuration of AI services where the network device can flexibly adjust AI service parameters, model information, and training data based on real-time network conditions and service requirements. This dynamic approach enables the system to adapt to changing demands without requiring complete reconfiguration.

Inventive Principle:
Principle #15Dynamics

2Adaptability or versatility

If comprehensive AI service configuration information is provided, then the AI computing functionality is improved, but the information transmission overhead increases

Engineering Contradiction:
ImproveAI service configuration capabilityVSAvoidconfiguration information volume
Core Design Contradiction:
Adaptability or versatilityVSQuantity of substance

Solution Approach 1:

The patent extracts essential AI service configuration elements from the overall configuration process and transmits them as separate information elements. By extracting only the necessary AI service identity, type, model information, and hyperparameters rather than complete system configurations, the patent reduces transmission overhead while maintaining comprehensive AI service capability.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent implements partial configuration transmission by providing only the specific AI service parameters that are currently needed (model information, training data, hyperparameters) rather than transmitting complete system configurations. This partial action approach reduces information volume while ensuring sufficient AI service functionality.

Inventive Principle:
Principle #16Partial or excessive action

3Manufacturing precision

If multiple AI service parameters are configured, then the AI computing precision and control are improved, but the configuration complexity increases

Engineering Contradiction:
ImproveAI service configuration precisionVSAvoidparameter management complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent divides AI service configuration into distinct segmented parameters including AI service type (training, inference, fine-tuning), model information, training data information, and hyperparameters. This segmentation allows precise control of each parameter independently while simplifying the overall management through structured organization.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by providing specific configuration precision only where needed - for example, detailed model information and hyperparameters are configured with high precision for AI computing tasks, while other less critical parameters use standard configurations. This targeted precision approach maintains AI computing accuracy without uniformly increasing overall complexity.

Inventive Principle:
Principle #3Local quality

Data Source

PatentUS20260089071A1Communication method and related device
Publication Date: 2026.03.26 HUAWEI TECH CO LTD
  • US20260089071A1 patent drawing
  • US20260089071A1 patent drawing
  • US20260089071A1 patent drawing

AI summary

This application provides a communication method and a related device, so that in a manner in which a network device provides configuration information of an AI service, a terminal device can transmit service information of the AI service based on the configuration information of the AI service, to implement integration of AI computing power and a communication network. In the method, the terminal device receives the configuration information of the AI service from the network device, where the configuration information of the AI service indicates configuration information of a connection function and at least one of the following: an identifier of the AI service, configuration information of a computing function, configuration information of a sensing function, and configuration information of a data function. The terminal device receives or sends the service information of the AI service based on the configuration information of the AI service.