Adaptive Numerology for Beamforming Training Overhead
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In multi-antenna networks, beamforming training overhead is high due to the need for multiple beam repetitions, which increases delay and resource consumption, especially in next-generation communication systems like 5G where high beamforming gain is required.
Innovation Solution
Implementing an adaptive numerology for beamforming training that configures radio resources with larger sub-carrier spacing (SCS) for OFDM symbols, allowing more training signals to be transmitted without additional antenna requirements, and dynamically adjusting SCS based on terminal device geometry, mobility, and link conditions to optimize beamforming training efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If beamforming training is performed with higher beamforming gain using more training beams, then beamforming accuracy is improved, but training overhead and delay increase
Solution Approach 1:
The patent applies dynamics by making the subcarrier spacing configurable and adaptable rather than fixed. The system dynamically adjusts the subcarrier spacing parameter for beamforming training signals based on channel conditions, allowing optimization between training accuracy and overhead. This dynamic parameter adjustment enables the system to adaptively select appropriate training configurations without requiring additional antenna elements or fixed resource allocations.
Solution Approach 2:
The core invention changes the numerology parameter (subcarrier spacing) specifically for beamforming training signals. By modifying this parameter, the system can transmit more training beams within the same time resources or reduce training time while maintaining accuracy. The patent explicitly configures different subcarrier spacing values for training signals versus data signals, enabling flexible control over training overhead and performance.
2Reliability
If more training beams are transmitted to achieve higher beamforming gain, then beamforming performance is improved, but resource consumption increases
Solution Approach 1:
The patent changes the subcarrier spacing parameter to increase the number of training beams that can be transmitted within available radio resources. By using a larger subcarrier spacing for training signals, more orthogonal resources become available, allowing more beams to be trained without proportionally increasing time or bandwidth consumption. This parameter change directly addresses the resource consumption issue while maintaining performance.
Solution Approach 2:
The patent introduces a new dimension for resource allocation by using different numerologies (subcarrier spacing values) specifically for training signals. This creates an additional degree of freedom in the time-frequency resource space, allowing the system to pack more training beams into the same physical resources by exploiting the enhanced orthogonality provided by larger subcarrier spacing.
3Device complexity
If traditional beamforming training is used without adaptive numerology, then system complexity is kept simple, but training overhead cannot be reduced
Solution Approach 1:
The patent introduces parameter changes (adaptive numerology) in a controlled manner that maintains system simplicity. Rather than redesigning the entire beamforming framework, the invention modifies only the subcarrier spacing parameter for training signals. This selective parameter adjustment reduces training overhead while adding minimal complexity, as the change is confined to specific signal processing aspects without requiring fundamental system restructuring.
Data Source
AI summary
A method for communications is proposed. The method may comprise determining a numerology for beamforming training supportable by a network node and a terminal device. The method may further comprise configuring the numerology for the beamforming training. The numerology may define radio resources with adaptive properties used for transmissions of beamforming training signals. The method may further comprise performing the beamforming training based at least partly on the configured numerology.