UE Handover Prediction Using Federated Location Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Handover procedures in communications networks require extensive inter-frequency measurements by user equipment (UE), leading to increased battery consumption and signaling overhead, particularly with the deployment of mmWave in 5G networks, and the use of target carrier prediction methods still necessitates frequent measurement of source carrier information, including sensitive location data.
Innovation Solution
Implementing a federated machine learning process to train a model on UE devices for predicting handover conditions based on location, allowing for fewer inter-frequency measurements and maintaining UE location privacy by keeping location data local.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If frequent inter-frequency measurements are performed to determine handover conditions, then measurement precision is improved, but battery power consumption increases
Solution Approach 1:
The system performs preliminary measurements and training during periods of low network load or using historical data, building up a database of handover conditions and UE characteristics before actual handover decisions are needed. This allows the UE to make informed handover decisions without performing frequent real-time inter-frequency measurements, thus reducing battery power consumption while maintaining prediction accuracy.
Solution Approach 2:
The patent replaces the mechanical process of frequent physical measurements with a computational model that uses machine learning algorithms. Instead of continuously measuring all frequency carriers, the system uses a trained model that predicts handover conditions based on processed historical data and UE characteristics, substituting physical measurement with computational prediction to reduce energy consumption.
2Productivity
If location data is used for handover decisions, then handover optimization is improved, but data privacy is compromised
Solution Approach 1:
The system extracts and processes only the necessary location-related features from the full location data, separating the useful handover optimization information from the sensitive privacy-critical components. By extracting only the minimal location features needed for handover decisions and processing them locally on the UE, the system maintains optimization efficiency while minimizing privacy exposure.
Solution Approach 2:
The patent introduces an intermediary processing layer that handles location data locally on the UE before any network interaction. The UE's local processor acts as an intermediary that enriches location data with UE-specific characteristics and performs handover decisions independently, preventing raw location data from being transmitted to the network while still achieving optimized handover decisions.
3Adaptability or versatility
If multiple receive chains are deployed to enable simultaneous multi-frequency measurements, then measurement capability is improved, but device complexity increases
Solution Approach 1:
The system performs preliminary measurements and characterizations during device initialization or low-activity periods, building up a database of frequency-specific characteristics and handover patterns. This preliminary data collection allows the UE to make adaptive handover decisions without requiring multiple simultaneous receive chains, reducing hardware complexity while maintaining multi-frequency adaptability.
Solution Approach 2:
Instead of physically implementing multiple receive chains for simultaneous measurements, the patent creates virtual copies of measurement capabilities through software-based signal processing and machine learning models. The UE uses a single receive chain to collect data and then employs computational models to simulate and predict multi-frequency conditions, achieving multi-frequency capability without the hardware complexity of multiple physical receive chains.
Data Source
AI summary
In a method in a user equipment, UE, in a communications network, of determining whether to perform a handover procedure from a first network node to a second network node, a location of the UE is provided as input to a model stored on the UE, the model having been trained using a machine learning process to predict conditions on the second network node in the communications network based on the location of the UE. A prediction of conditions on the second network node at the provided location of the UE is provided by the model. The received predicted conditions are then used to determine whether to perform a handover procedure.