Self-learning UE Capability Verification from Partial Reports
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In self-learning user equipment (UE) capability verification methods, existing technologies face challenges in handling partial capabilities reported by UEs in different geographical areas, leading to inefficiencies in building a complete UE capability set due to the exponential increase in RF bands and band combinations in 4G and 5G networks.
Innovation Solution
A self-learning algorithm is employed to merge disjointed band combination information from UEs in different geographical areas, considering partial UE capabilities and geographical area knowledge to build a complete UE capability set, with optional band and feature filters applied to validate and enhance the capability verification process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the network requests complete UE capability reports from all UEs, then the capability verification accuracy is improved, but the signaling overhead and network traffic increase exponentially due to the large number of RF bands and band combinations in 4G and 5G networks
Solution Approach 1:
The patent extracts only the necessary capability information from UE reports by identifying and filtering relevant band combinations based on network deployment knowledge. Instead of processing complete capability reports, the system extracts minimal sufficient information to verify UE capabilities, thereby reducing signaling overhead while maintaining verification accuracy.
Solution Approach 2:
The patent applies local quality by customizing capability verification for different geographical areas and network deployments. Each network node uses its specific knowledge of local band combinations and UE capabilities to process reports selectively, rather than applying a uniform approach to all UEs, thus reducing overall signaling overhead while maintaining local verification accuracy.
2Loss of information
If the network stores complete capability sets for all UEs, then the capability information completeness is improved, but the memory requirements and processing complexity increase exponentially with the number of RF bands and band combinations
Solution Approach 1:
The patent segments the capability verification process into multiple stages: initial capability reporting, network knowledge matching, and selective verification. By dividing the process into segments, the system avoids the need to store and process complete capability sets for all UEs simultaneously, reducing memory requirements and processing complexity while maintaining information completeness.
Solution Approach 2:
The patent performs preliminary action by pre-storing network knowledge about band combinations and UE capabilities in a database before actual capability verification is needed. This preliminary preparation allows the system to quickly match and verify UE capabilities without complex real-time processing, reducing both memory requirements and processing complexity during operation.
3Measurement precision
If the self-learning algorithm processes all reported capabilities from UEs, then the capability dictionary accuracy is improved, but the learning time and convergence speed deteriorate due to the exponential increase in capability data volume
Solution Approach 1:
The patent applies partial action by processing only the necessary subset of capability reports needed to build an accurate capability dictionary. Instead of processing all reported capabilities from all UEs, the system selectively processes reports that provide new or corrective information, thereby achieving dictionary accuracy with reduced learning time.
Solution Approach 2:
The patent implements feedback mechanisms where the capability dictionary is continuously refined based on incoming UE reports and network observations. The system uses feedback from capability verification results to update and improve the dictionary accuracy over time, rather than requiring complete processing of all possible capability data, thus reducing learning time while maintaining accuracy.
Data Source
AI summary
An apparatus of a node of a network comprises one or more baseband processors to process a first user equipment (UE) capability report from a first UE, and to process a second UE capability report from a second UE, wherein the first UE capability report and the second UE capability report include a same UE capability ID for the first UE and the second UE, and wherein the first capability report includes partial UE capability information for the first UE and the second capability report includes partial UE capability information for the second UE. The apparatus can include a memory to store the first capability report and the second capability report.