Communication Network AI Model Switching for Channel-Aware Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing AI models in wireless cellular network-based positioning technologies face challenges in maintaining positioning precision due to changes in channel environments, leading to deteriorated performance.
Innovation Solution
A communication method and apparatus that determine whether to switch or update AI models for positioning based on channel measurement results, using a correspondence between AI models and specific parameters to adapt to environmental changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single AI model is used for positioning, then device complexity is reduced, but positioning precision deteriorates due to channel environment changes
Solution Approach 1:
The patent implements dynamic AI model switching based on channel environment conditions. The system monitors channel measurements (such as signal strength, latency, packet loss) and dynamically selects the most appropriate AI model from multiple pre-trained models, allowing the positioning system to adapt to changing environmental conditions without requiring a single complex omniscient model.
Solution Approach 2:
The patent changes the parameter of AI model selection based on channel environment parameters. By monitoring channel quality indicators and switching between different AI models according to these parameters, the system optimizes positioning precision for different channel conditions while maintaining manageable complexity through parameter-based decision making.
2Measurement precision
If AI model switching is implemented to adapt to channel changes, then positioning precision is improved, but device complexity increases
Solution Approach 1:
The patent performs preliminary actions by pre-training multiple AI models for different channel environments and pre-establishing a selection mechanism. This allows the system to avoid the complexity of training and selecting models in real-time, as the models are prepared in advance and selection is based on straightforward channel condition matching.
Solution Approach 2:
The patent implements feedback mechanisms where channel measurements are continuously monitored and fed back to the model selection process. This feedback loop enables the system to adjust AI model selection based on actual channel conditions, improving positioning precision while keeping the control logic simple through condition-based decision trees.
Data Source
AI summary
This application provides a communication method, including: A network element obtains a measurement result of a first parameter based on a channel measurement result, and determines, based on a correspondence between the first parameter and an AI model, a specific AI model corresponding to the current measurement result of the first parameter. The AI model corresponding to the measurement result of the first parameter is compared with an AI model currently used for positioning, to determine whether to switch or update the AI model used for positioning. In this way, determining, based on the measurement result of the first parameter, whether to switch or update the AI model can implement switching or updating of the AI model in a timely manner based on the change of the channel environment, thereby improving positioning precision of the AI model.