AI Model Information Transmission Using Segmented Resource Configuration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge of efficiently transmitting information related to AI/ML models between devices, particularly from network devices to terminal devices with limited resources, is not adequately addressed in existing communication systems.
Innovation Solution
A method and apparatus for transmitting AI information using a transmission resource configuration, including model identity, description, and algorithm data, facilitated by a first device that receives and sends AI information using configurations provided by a second device, utilizing radio bearers, session identities, and data importance levels to ensure accurate reassembly and delivery.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If AI information is transmitted using existing communication systems, then transmission can occur, but resource utilization is inefficient and transmission reliability is insufficient for terminal devices with limited resources
Solution Approach 1:
The patent segments AI information transmission into multiple parts including model identity information, model description information, and model algorithm data information. Each segment is transmitted using different resource configurations and can be reassembled at the receiving end, improving both efficiency and reliability for devices with limited resources
Solution Approach 2:
The patent introduces a network device as an intermediary that facilitates AI information transmission between terminal devices. The network device manages resource configurations, coordinates bidirectional transmission, and ensures reliable delivery, resolving the contradiction between efficient resource utilization and transmission reliability
2Adaptability or versatility
If transmission resources are allocated for AI information, then transmission capability is improved, but resource complexity increases for management and configuration
Solution Approach 1:
The patent creates a universal resource configuration framework that can handle multiple types of AI information (model identity, description, algorithm data) using a standardized configuration structure. This multi-functional approach improves transmission capability while reducing configuration complexity through reuse of the same framework across different information types
3Adaptability or versatility
If bidirectional AI information sharing is enabled, then model management flexibility is improved, but transmission overhead and resource consumption increase
Solution Approach 1:
The patent implements dynamic resource configuration for bidirectional AI information sharing, where resource allocation is adjusted based on actual transmission needs, device capabilities, and information importance. This dynamic approach maintains model management flexibility while optimizing resource consumption by allocating resources only when and where needed
Data Source
AI summary
An information transmission method and a device are provided. The method includes the following. The first device transmits artificial intelligence (AI) information by using a transmission resource configuration sent by a second device, where the AI information is information related to an AI/machine learning (ML) model.