Subspace Precoding From Denoised SRS for Wideband MU-MIMO
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 5G/NR communication systems face challenges in achieving efficient radio interface coverage and capacity due to limited CSI-RS ports and noisy sounding reference signals (SRS) affecting multi-user precoding, especially in high-frequency bands and distributed MIMO configurations.
Innovation Solution
Implementing subspace precoding techniques at base stations to derive a common precoder for multiple resource blocks based on denoised SRS from user equipments, providing a wideband multi-user precoding scheme.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional CSI-RS based precoding is used, then system complexity is reduced, but precoding accuracy deteriorates due to limited CSI-RS ports
Solution Approach 1:
The patent introduces SRS (Sounding Reference Signal) as an intermediary to enable uplink channel estimation for precoding purposes. By using SRS transmitted by UEs, the base station can obtain channel state information without relying on limited CSI-RS ports, thus improving precoding accuracy while avoiding the complexity of extensive CSI-RS configurations
Solution Approach 2:
The patent employs channel reciprocity to copy downlink channel characteristics from uplink SRS measurements. The base station transmits downlink reference signals, UEs estimate the downlink channel and feed back SRS, allowing the base station to derive downlink precoding information from uplink measurements, effectively copying channel state information across directions
2Loss of information
If wideband precoding is applied, then signaling overhead is reduced, but adaptation to frequency-selective fading deteriorates
Solution Approach 1:
The patent segments the frequency band into multiple resource block groups (RBGs), deriving separate precoders for each RBG based on SRS measurements. This allows the system to maintain a wideband precoder for overall structure while applying frequency-selective adjustments at the RBG level, balancing signaling overhead reduction with frequency adaptation capability
Solution Approach 2:
The patent implements dynamic precoder selection where the base station can switch between wideband precoding and frequency-selective precoding based on channel conditions. When channel variations are mild, wideband precoding reduces overhead; when frequency-selective fading is severe, the system activates RBG-level precoders to adapt to frequency variations
3Productivity
If multi-user MIMO is implemented, then system capacity increases, but performance deteriorates under noisy SRS conditions
Solution Approach 1:
The patent applies preliminary denoising processing to SRS signals before using them for multi-user precoding. By filtering and cleaning the SRS measurements in advance, the system removes noise components that would otherwise degrade precoding accuracy, ensuring reliable MU-MIMO performance even in noisy environments
Solution Approach 2:
The patent implements feedback mechanisms where UEs report channel quality indicators and precoding matrix indicators based on received reference signals. The base station uses this feedback to adjust precoding strategies, selecting appropriate precoders from codebooks that are robust to noise, thereby maintaining MU-MIMO capacity under noisy SRS conditions
Data Source
AI summary
A base station (BS) includes a transceiver. The transceiver is configured to receive, from k user equipments (UEs), a sounding reference signal (SRS), wherein k is an integer greater than or equal to 1. The BS also includes a processor operatively coupled to the transceiver. The processor is configured to denoise the SRS received from the k UEs, derive, based on the denoised SRS from the k UEs, a common precoder for a plurality of resource blocks (RBs), as a wideband precoder, and provide a wideband multi-user precoding scheme for the k UEs based on the wideband precoder.