Terminal Measurement Configuration for AI Model Training Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing AI models in communication systems face challenges due to limited storage space and insufficient data collection, leading to ineffective training, especially in scenarios with different engineering parameters and antenna forms.
Innovation Solution
A communication method and apparatus that involves configuring terminal devices with targeted measurement information to collect and report data for AI model training, ensuring efficient data collection and secure data transfer between network devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the terminal device collects excessive measurement data, then the AI model training data sufficiency is improved, but the terminal device storage capacity is exceeded
Solution Approach 1:
The patent extracts only the necessary measurement data elements required for AI model training by configuring specific measurement parameters and data types through network device instructions. The terminal device collects data according to pre-defined configuration information that specifies what data to measure and report, avoiding collection of unnecessary data that would consume storage space.
Solution Approach 2:
The patent changes the parameters of data collection by allowing the network device to dynamically configure measurement parameters, data types, and reporting frequencies. This enables the system to adjust the quantity and quality of collected data according to specific training requirements, optimizing the balance between data sufficiency and storage constraints.
2Volume of stationary object
If the terminal device collects insufficient measurement data, then the terminal device storage is preserved, but the AI model training effectiveness deteriorates
Solution Approach 1:
The patent implements a feedback mechanism where the network device receives measurement data from the terminal device and uses it to evaluate AI model training progress. Based on this feedback, the network device can adjust configuration information to request additional or different measurement data, ensuring that sufficient quality data is collected for effective model training while avoiding unnecessary data collection.
Solution Approach 2:
The patent performs preliminary action by pre-configuring the terminal device with measurement parameters and data types before data collection begins. The network device determines the optimal measurement configuration in advance based on AI model training requirements, so the terminal device collects exactly the right data from the start, avoiding both data deficiency and excessive storage usage.
3Measurement precision
If the terminal device performs comprehensive measurement, then the measurement data quality is improved, but the measurement complexity increases
Solution Approach 1:
The patent introduces the network device as an intermediary that manages measurement configuration complexity. The network device receives AI model training requirements, determines appropriate measurement parameters and data types, and translates these into configuration information for the terminal device. This intermediary role simplifies the terminal device's task while ensuring high-quality data collection according to training needs.
Data Source
AI summary
Embodiments of this application provide a communication method and apparatus. The method is applied to model training in the artificial intelligence field. The method includes: A terminal device receives configuration information from a first network device, performs measurement based on the configuration information, to obtain measurement data associated with an AI model, and sends the measurement data to the first network device, so that the first network device can complete training of the AI model based on the measurement data; or the terminal device completes training of the AI model based on the measurement data, and reports a trained model to the first network device. The configuration information is determined based on an optimization requirement for training the AI model of a second network device. According to the solutions of this application, the terminal device can obtain valid measurement data, to support AI model training and improve model training effect.