UE Radio Frequency Fingerprint Positioning With TRE Prioritization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless positioning technologies in 5G networks face challenges in accurately determining the location of user equipment due to the complexity of radio frequency environments, leading to inefficiencies in selecting and prioritizing transmission/reception entities for radio frequency fingerprint-based positioning.
Innovation Solution
A method involving user equipment (UE) and network entities that utilize machine learning models to acquire, select, and prioritize transmission/reception entities for radio frequency fingerprint (RFFP) measurements, enabling precise UE positioning through trained models and criteria for selecting and prioritizing TREs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional wireless positioning methods are used in 5G networks, then positioning can be performed, but positioning accuracy deteriorates due to radio frequency environment complexity
Solution Approach 1:
The patent introduces transmission/reception entity (TRE) selection as an intermediary mechanism between the UE and the positioning system. By selecting specific TREs based on signal quality metrics (RSRP, RSRQ, SINR) and geometric criteria (GDOP, PDOP, HDOP), the system mediates the complex RF environment to provide cleaner, more reliable positioning data, thereby improving positioning accuracy without requiring the UE to directly process all environmental complexities
Solution Approach 2:
The patent replaces traditional geometric positioning methods with radio frequency fingerprinting (RFFP) based positioning. Instead of relying solely on geometric relationships between TREs and UE, the system uses RF signal characteristics (fingerprints) as the primary positioning mechanism. This substitution allows the system to account for RF environment complexities by comparing measured signal characteristics against pre-collected fingerprint databases, thereby improving positioning accuracy in complex RF environments
2Quantity of substance
If all transmission/reception entities are used for positioning measurements, then measurement coverage is maximized, but processing time and computational load increase
Solution Approach 1:
The patent extracts and selects only the most relevant TREs from the complete set of available TREs. By applying selection criteria that evaluate signal quality (RSRP, RSRQ, SINR thresholds) and geometric quality (GDOP, PDOP, HDOP thresholds), the system extracts a subset of high-quality TREs for positioning measurements. This extraction process reduces the number of TREs that need to be processed while maintaining positioning accuracy, thereby reducing processing time and computational load
Solution Approach 2:
The patent applies partial action by using only the necessary number of TREs for accurate positioning rather than all available TREs. The selective TRE identification and measurement process performs partial measurements on a curated subset of TREs that meet quality thresholds, avoiding the excessive processing that would result from measuring all TREs. This partial approach maintains positioning accuracy while significantly reducing processing time
3Reliability
If radio frequency fingerprinting measurements are performed from all TREs, then positioning data completeness is improved, but measurement complexity and processing overhead increase
Solution Approach 1:
The patent performs preliminary TRE selection and prioritization before conducting RFFP measurements. By pre-evaluating TREs based on signal quality metrics and geometric criteria, and by prioritizing TREs that are more likely to provide reliable measurements, the system prepares a curated list of TREs for RFFP measurement. This preliminary action ensures that measurements are performed on high-quality TREs, improving data completeness and reliability while reducing measurement complexity by avoiding low-quality TREs
Data Source
AI summary
Disclosed are techniques for UE-based wireless positioning. In an aspect, a user equipment (UE) may acquire, from a first set of transmission/reception entities (TREs), a first set of measurements of signals. The UE may select, from the first set of TREs, a second set of TREs, based on criteria for selecting and/or prioritizing TREs for radio frequency fingerprint (RFFP)-based positioning. The UE may acquire, from the second set of TREs, a second set of measurements, the second set of measurements comprising RFFP measurements. The UE may estimate a position of the UE based on the second set of measurements. In an aspect, the UE may input the second set of measurements into a trained machine learning (ML) model that outputs an estimated position of the UE. In an aspect, the UE receives the selection/prioritization criteria and/or the trained ML model from a network node.