UE Historical Movement Reporting for Accurate Cell Handover Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current cell handover processes in wireless communication systems rely solely on signal strength, lacking sufficient movement information for accurate machine learning-based predictions when the user equipment (UE) is not in a connected state, leading to incomplete training and inaccurate predictions.
Innovation Solution
The UE records and reports historical movement information, including cell state and change information, to the base station, ensuring comprehensive data collection for machine learning models, even when not in a connected state.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the UE does not report historical movement information, then the device complexity is reduced, but the completeness of training data for machine learning models deteriorates
Solution Approach 1:
The UE performs preliminary recording of movement information (cell ID, timestamp, location) in its memory before any machine learning training occurs. This preliminary action ensures that when the UE later enters a connected state, the base station can immediately obtain comprehensive historical movement data without requiring continuous real-time reporting, thus resolving the contradiction between reducing reporting complexity and maintaining data completeness.
2Loss of information
If the UE reports all historical movement information continuously, then the completeness of training data is improved, but the loss of energy increases
Solution Approach 1:
Instead of continuous reporting, the UE adopts periodic action by only reporting historical movement information when specific conditions are met (e.g., when entering a connected state or when requested by the base station). This periodic reporting mechanism ensures that complete training data is provided while minimizing energy consumption by avoiding unnecessary continuous transmissions.
3Productivity
If machine learning training is performed without historical movement information, then the productivity is improved (faster training initiation), but the measurement precision deteriorates (prediction accuracy)
Solution Approach 1:
The UE performs preliminary recording of movement information in advance, storing cell ID, timestamp, and location data in its memory. This preliminary action allows the base station to immediately initiate machine learning training as soon as the UE enters a connected state, without waiting for continuous real-time data collection. The pre-recorded historical data provides both the speed for immediate training initiation and the completeness for accurate prediction, resolving the contradiction between productivity and measurement precision.
Data Source
AI summary
An information transmission method and apparatus, a communication device, and a storage medium are provided. The method includes that: a user equipment (UE) reports historical mobile information to a base station, where the historical mobile information includes first mobile information. Furthermore, the first mobile information is recorded by the UE in response to a cell change of the UE, and the first mobile information is associated with the cell change of the UE.