Computing Power Reconfiguration for AI Task Continuity in Handover

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing wireless networks struggle to ensure continuity of computing task execution during cell handovers, particularly in scenarios involving artificial intelligence, where both computing power and connection services are required.

Innovation Solution

A computing power reconfiguration method is implemented by a Computing Management Function (CMF) to obtain and utilize computing power information of target access network devices and core network elements, allowing for flexible migration of computing tasks and ensuring continuity during cell handovers.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If computing tasks are migrated during cell handover, then continuity of computing task execution is improved, but device complexity and network coordination overhead increase

Engineering Contradiction:
Improvecontinuity of computing task executionVSAvoidnetwork coordination overhead
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The CMF obtains computing power information of target access network devices in advance before handover occurs. This preliminary action allows the system to pre-evaluate suitable target devices for computing task migration, reducing the coordination overhead during actual handover while ensuring continuity of computing task execution.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The CMF acts as an intermediary between access network devices and core network elements. It centrally manages computing power information and coordinates migration decisions, simplifying the overall network coordination by providing a single point of control rather than requiring direct peer-to-peer communication between multiple network entities.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Adaptability or versatility

If computing power information is collected and processed, then computing task migration capability is improved, but information processing overhead and signaling traffic increase

Engineering Contradiction:
Improvecomputing task migration capabilityVSAvoidsignaling traffic overhead
Core Design Contradiction:
Adaptability or versatilityVSLoss of information

Solution Approach 1:

The computing power information obtained by the CMF serves multiple purposes: it enables computing task migration decisions, supports service quality maintenance, and facilitates continuity assurance during handover. This multi-functionality reduces the need for separate information collection processes, thereby reducing overall signaling traffic overhead.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

Access network devices and core network elements provide computing power information to the CMF in a self-service manner as part of their normal operations. This reduces the need for additional dedicated signaling mechanisms, as the information is obtained through existing network communication channels during regular handover procedures.

Inventive Principle:
Principle #25Self-service

3Reliability

If computing resources are dynamically reallocated, then service quality during handover is improved, but control complexity and processing time increase

Engineering Contradiction:
Improveservice quality during handoverVSAvoidcontrol complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system dynamically adjusts computing task allocation based on changes in computing power information parameters of access network devices. By monitoring and responding to parameter changes (such as available computing resources, device capacity), the CMF can reallocate computing resources efficiently without requiring complex control logic, thereby maintaining service quality while managing control complexity.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS20260006550A1Computing power reconfiguration method and related apparatus
Publication Date: 2026.01.01 HUAWEI TECH CO LTD
  • US20260006550A1 patent drawing
  • US20260006550A1 patent drawing
  • US20260006550A1 patent drawing

AI summary

receiving an acknowledgment message for the CE configuration message from the target access network device, and receiving an acknowledgment message for the CE reconfiguration or deletion message from the first core network element. According to embodiments of this application, continuity of computing task execution can be implemented in a cell handover scenario.