Implicit CSI Delivery via Correlation Matrix for Massive MIMO Fronthaul
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional CSI delivery methods in massive MIMO systems consume significant fronthaul bandwidth due to the large amount of CSI-related information that needs to be transmitted from the RRU to the BBU, which becomes a bottleneck as the size of the antenna array increases.
Innovation Solution
Implementing an implicit CSI delivery scheme by transmitting the correlation-relationship matrix (HH^H) instead of the explicit CSI matrix (H), which does not scale with the size of the antenna array, and using a two-part precoder design to reduce the amount of precoding information transmitted.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all CSI of each UE is delivered from the RRU to the BBU, then the BBU can perform MIMO scheduling and precoding, but the fronthaul bandwidth consumption increases significantly with the size of the antenna array
Solution Approach 1:
The patent extracts only the essential information needed for MIMO scheduling and precoding by delivering the correlation-relationship matrix (HH^H) instead of the complete CSI matrix H. This extraction principle reduces the information volume from N*K elements to K*K elements while preserving the necessary statistical properties for beamforming and scheduling decisions at the BBU.
Solution Approach 2:
The patent transforms the CSI representation from explicit channel coefficients to implicit correlation statistics. By changing the parameter form from H to HH^H, the system maintains the essential spatial correlation information needed for precoding while significantly reducing the data dimension, achieving bandwidth reduction without sacrificing scheduling performance.
2Productivity
If the size of the antenna array increases in massive MIMO systems, then the system capacity and spectral efficiency improve, but the amount of CSI-related information to be delivered increases significantly
Solution Approach 1:
The patent extracts the essential statistical characteristics of the channel (correlation matrix) rather than transmitting the full channel state information. This allows the system to support larger antenna arrays in massive MIMO configurations because the extracted correlation information captures the dominant spatial characteristics without scaling with the number of antennas N.
Solution Approach 2:
The patent uses the correlation-relationship matrix HH^H as a statistical copy or representation of the full CSI matrix H. This copied representation preserves the essential spatial correlation structure needed for MIMO processing while being independent of the antenna array size N, enabling scalable massive MIMO implementations.
3Quantity of substance
If a two-part precoder design is used, then the amount of precoding information transmitted is reduced, but the complexity of precoder construction increases
Solution Approach 1:
The patent segments the precoder construction into two parts: a first precoder matrix W1 that is determined at the RRU based on local channel statistics, and a second precoder matrix W2 that is determined at the BBU based on received correlation information. This segmentation distributes the computational complexity across both units while reducing the information that must be transmitted over the fronthaul link.
Solution Approach 2:
The patent performs preliminary processing of the channel information at the RRU by computing the correlation matrix HH^H and determining the first precoder W1 before transmission to the BBU. This preliminary action reduces the information volume that needs to be transmitted and allows the BBU to focus on determining only the second precoder component W2, simplifying the overall precoder construction process.
Data Source
AI summary
This disclosure relates to a method for providing channel state information (CSI) from a remote radio unit (RRU) to a baseband unit (BBU), the method comprising: determining CSI for each user equipment (UE) of a plurality of UEs based on a reference signal received from the respective UE; generating a plurality of correlation coefficients based on the CSI; and providing the plurality of correlation coefficients to the BBU.