Precoding Over Beam Subset for NLoS Channels
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Massive beamforming in non-line-of-sight (NLoS) channels faces challenges such as reduced beamforming gain due to large angular spread and excessive feedback overhead, particularly in systems with large antenna arrays.
Innovation Solution
A method for precoding over a beam subset, where a network node exchanges reference signal information with a wireless device to determine a proper subset of transmission beams and calculates precoding weights, reducing the number of transmission beams while maintaining high beamforming gain and minimizing feedback overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional beamforming is used in NLoS channels with large antenna arrays, then beamforming gain is reduced due to large angular spread, but using MRT precoding requires excessive feedback overhead
Solution Approach 1:
The patent segments the full set of antenna elements into multiple beam groups, where each beam group corresponds to a subset of antenna elements. This segmentation allows the system to perform precoding on a reduced basis set of beams rather than all antenna elements, thereby reducing feedback overhead while maintaining beamforming gain through coherent combination of selected beams.
Solution Approach 2:
The patent extracts and utilizes only the dominant beam components from the channel response by identifying the strongest beams through reference signal measurements. By taking out only the essential beam information needed for precoding rather than requiring full channel state information, the system achieves effective beamforming with significantly reduced feedback requirements.
2Reliability
If full-dimensional digital beamforming is used to achieve high beamforming gain, then communication performance is improved, but baseband and radio resource requirements increase significantly
Solution Approach 1:
The patent divides the large antenna array into multiple beam groups, each handled by a separate precoding operation. This segmentation reduces the computational complexity at the baseband by working with smaller subsets of antenna elements independently, while the radio resources are optimized by transmitting reference signals only for the selected beam groups rather than all antenna elements.
Solution Approach 2:
The patent applies partial action by using only a subset of beams (rather than all antenna elements) for precoding. This partial approach achieves sufficient beamforming gain for practical communication scenarios without requiring the full computational and resource investment of complete digital beamforming across all antenna elements.
Data Source
Figure 1a~2b
Figure 3~6
Figure 5
AI summary
There is provided mechanisms for precoding over a beam subset. A method is performed by a network node. The method comprises exchanging reference signal information (S102) for a set of transmission beams or set of antenna ports with a wireless device. The method comprises acquiring (S104), based on the reference signal information, information regarding which proper subset of transmission beams from the set of transmission beams to use for communication with the wireless device. The method comprises determining precoding weights (S106) for the proper subset of transmission beams or antenna ports based on the reference signal information.