MIMO Channel Reconstruction from Coarse PMI and Sparse CSI-RS
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communication systems face challenges in reducing quantization noise and resource overhead due to large quantization granularity in Precoding Matrix Indication (PMI) information and Demodulation Reference Signal (DMRS) density, which affects Multiple-Input Multiple-Output (MIMO) precoding and channel reconstruction accuracy.
Innovation Solution
Implementing a base station and terminal configuration that uses channel state information reference signals with reduced density and finer granularities, combined with super-resolution networks for interpolation and denoising, to reconstruct channels with improved precision while minimizing signaling overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the granularity of PMI information is improved, then channel reconstruction precision is improved, but resource overhead of CSI-RS increases
Solution Approach 1:
A super-resolution network is introduced as an intermediary between the low-density CSI-RS and the channel reconstruction process. The network processes the limited reference signal data to generate high-precision channel estimates, effectively mediating between the coarse input and the fine output requirements
Solution Approach 2:
The system changes the parameter of CSI-RS density from high to low, while compensating for the reduced measurement density through the super-resolution network's signal processing capabilities. This parameter transformation allows maintaining channel estimation accuracy while reducing reference signal overhead
2Measurement precision
If the density of DMRS is improved, then channel estimation precision is improved, but resource overhead of reference signal increases
Solution Approach 1:
The super-resolution network serves as an intermediary that enhances the effectiveness of low-density DMRS. By processing the limited reference signal samples through learned signal models, the system achieves high-precision channel estimation without increasing DMRS density
Solution Approach 2:
The patent replaces the traditional mechanical approach of increasing reference signal density with a signal processing-based solution. Instead of adding more physical reference signals, the system uses computational methods (super-resolution networking) to extract more information from existing signals
3Loss of substance
If subband-level PMI information is used, then signaling overhead is reduced, but quantization noise increases
Solution Approach 1:
The system converts the limitation of subband-level PMI granularity into a benefit by using the super-resolution network to compensate for the quantization effects. The network learns to reconstruct fine-grained channel characteristics from coarse subband PMI information, turning the quantization limitation into an opportunity for demonstrating advanced signal processing capabilities
Data Source
AI summary
The present disclosure provides a base station, a terminal and channel reconstruction methods performed by a base station and a terminal. The base station includes: a transmitting unit configured to transmit channel state information reference signals of multiple ports to a terminal, wherein the channel state information reference signals of multiple ports have a first density in frequency domain in one time interval; a receiving unit configured to receive precoding matrix indication information of a first granularity from the terminal; a processing unit configured to determine a channel of a second granularity according to the precoding matrix indication information of the first granularity, and perform downlink precoding on the channel, wherein the second granularity is finer than the first granularity.