Beam Management Using ML Candidate Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current beam management in wireless communication systems faces challenges in efficiently managing candidate beams without separate signaling, leading to increased beam tracking latency and resource overhead, particularly in scenarios with high mobility or frequent beam disconnections.
Innovation Solution
A method using machine learning to determine and manage optimal candidate beams by receiving configuration information from a base station, including resource information and candidate beam determination algorithms, allowing the user equipment to measure and report only necessary beams, thereby reducing unnecessary signaling and resource allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional beam management methods are used with separate signaling for each candidate beam, then beam tracking accuracy is maintained, but beam tracking latency increases and resource overhead increases
Solution Approach 1:
The patent segments the beam management process by dividing candidate beams into groups based on measurement results. Instead of managing all candidate beams individually with separate signaling, the system segments them into selected and non-selected groups, applying different handling strategies to reduce latency while maintaining accuracy for the most promising beams.
Solution Approach 2:
The patent extracts only the most relevant candidate beams for detailed management. By identifying and extracting the top N candidate beams based on measurement results, the system focuses resources on the most promising options, reducing overall beam tracking latency while maintaining accuracy for the selected beams.
2Reliability
If separate signaling is used for each candidate beam, then beam management completeness is ensured, but resource overhead increases
Solution Approach 1:
The patent extracts only the essential candidate beams that require separate signaling management. By identifying the top N candidate beams based on measurement results and applying separate signaling only to these extracted beams, the system ensures completeness for the most relevant options while significantly reducing resource overhead compared to managing all candidate beams individually.
Solution Approach 2:
The patent applies partial action by managing only a subset of candidate beams with detailed separate signaling. Instead of applying complete signaling management to all candidate beams, the system applies partial management to the top N beams, which provides sufficient reliability for beam tracking while reducing resource overhead.
3Measurement precision
If all candidate beams are measured and managed, then beam selection accuracy is improved, but power consumption in user equipment increases
Solution Approach 1:
The patent extracts only the most relevant candidate beams for detailed measurement and management. By identifying and extracting the top N candidate beams based on initial measurements, the system performs detailed beam management only on these extracted beams, reducing power consumption while maintaining selection accuracy for the most promising beams.
Solution Approach 2:
The patent applies partial measurement action by measuring and managing only a subset of candidate beams in detail. Instead of measuring all candidate beams with equal precision, the system performs partial detailed measurement on the top N beams, which reduces power consumption while maintaining sufficient accuracy for beam selection.
Data Source
AI summary
A method of performing a beam management in a wireless communication system and a device therefor are disclosed. The method performed by a user equipment (UE) comprises receiving, from a base station (BS), configuration information including i) resource information on a plurality of reference signals and ii) candidate beam determination algorithm information, determining N candidate beams for at least one serving beam based on the candidate beam determination algorithm information, the N candidate beams being related to N reference signals of the plurality of reference signals, receiving, from the BS, the N reference signals of the plurality of reference signals based on the resource information on the plurality of reference signals, and transmitting, to the BS, measurement information including at least one measurement value among the N reference signals.