Downlink Precoding Using Delay-Spatial Reciprocity in FDD
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 5G NR precoding methods, such as Rel-15 and Rel-16 type II codebooks, suffer from significant feedback overhead and performance loss due to quantization errors, particularly in frequency division duplex (FDD) systems, as they do not effectively leverage both spatial and delay reciprocity in uplink and downlink channels.
Innovation Solution
Implement a partial reciprocity-based CSI framework that incorporates efficient and flexible designs for downlink reference signal precoders, utilizing uplink reference signals to estimate delay and spatial information at the gNB, and optimizes uplink control information format to reduce reporting overhead, enabling the UE to report only non-zero channel coefficients without needing to specify frequency domain subsets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If codebook-based precoding with type II codebooks is used to improve precoding accuracy, then PMI accuracy is improved, but feedback overhead increases significantly
Solution Approach 1:
The patent extracts only the essential channel information (non-zero channel coefficients) that is needed for accurate precoding, eliminating redundant feedback information. By identifying and reporting only the significant channel coefficients rather than complete channel state information, the system achieves accurate precoding with reduced feedback overhead.
Solution Approach 2:
The patent applies partial action by reporting only the non-zero channel coefficients instead of all channel coefficients. This partial reporting approach provides sufficient information for accurate precoding while significantly reducing the amount of feedback data transmitted, thus lowering overhead.
2Measurement precision
If full channel state information is reported to achieve accurate precoding, then precoding accuracy is improved, but uplink resource consumption increases
Solution Approach 1:
The patent extracts only the necessary channel information (non-zero coefficients) required for accurate precoding, removing redundant information. This extraction approach maintains precoding accuracy while minimizing uplink resource consumption by transmitting only essential data.
Solution Approach 2:
The patent uses partial action by reporting only the non-zero channel coefficients rather than complete channel state information. This partial reporting provides sufficient accuracy for precoding while significantly reducing uplink resource usage.
3Loss of information
If quantization is applied to reduce feedback data size, then feedback overhead is reduced, but quantization errors increase causing performance loss
Solution Approach 1:
The patent extracts only the non-zero channel coefficients for reporting, avoiding quantization of all channel coefficients. This selective extraction maintains precision by reporting exact non-zero values while reducing overhead by omitting zero or negligible coefficients.
Data Source
AI summary
There is provided a method at a network node of a communication network, comprising: receiving an uplink reference signal from a user equipment; estimating, based on the received uplink reference signal, delay information regarding a delay profile of a communication channel between the user equipment and the network node and spatial information regarding the communication channel; applying both the delay information and the spatial information in precoding of at least one downlink reference signal; transmitting the at least one precoded downlink reference signal to the user equipment; and receiving, from the user equipment as a response to the transmission of the at least one precoded downlink reference signal, channel information, the channel information indicating at least non-zero coefficients of at least one channel transmission layer.