Massive MIMO Uplink Signal Detection Using SRS Channel Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Massive MIMO systems face challenges in uplink signal detection due to the high computational requirements and processing delays, which are exacerbated by the large number of antennas and user equipment in conventional methods, leading to unacceptable latency and increased costs.
Innovation Solution
A new uplink signal detection process is implemented in massive MIMO systems, where channel coefficients are estimated using Sounding Reference Signals (SRS), and frequency and time offsets are compensated to compute effective channel vectors, allowing for the calculation of detection matrices that separate data signals from multiple user equipment, reducing processing delays and costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional uplink signal detection methods are used in massive MIMO systems, then channel coefficients can be estimated for data demodulation, but the computation of detection matrices requires huge hardware resources and causes unacceptable processing delay
Solution Approach 1:
The patent segments the uplink signal detection process into two distinct phases: (1) channel estimation phase using Sounding Reference Signals (SRS) where the BS estimates channel coefficients for multiple UEs, and (2) data detection phase using Demodulation Reference Signals (DMRS) where only the necessary detection matrices are computed. This segmentation allows channel information to be acquired in advance without computing full detection matrices for all resource blocks, thereby reducing processing delay while maintaining estimation accuracy.
Solution Approach 2:
The patent implements preliminary channel estimation using SRS transmitted in uplink subframes before data transmission. The BS computes channel coefficients in advance for multiple UEs across the entire bandwidth, storing these estimates for subsequent data detection. This preliminary action eliminates the need to compute detection matrices for every resource block during data reception, significantly reducing real-time processing requirements and delay.
2Productivity
If the number of receiving antennas and multiplexed UEs is increased to hundreds and more than ten respectively, then spatial multiplexing capability and array gains are improved, but the computation of detection matrices requires huge hardware resources and increases cost
Solution Approach 1:
The patent applies partial action by computing detection matrices selectively only for resource blocks containing data from UEs that require demodulation, rather than computing matrices for all possible UE combinations across the entire bandwidth. The BS uses the pre-acquired channel estimates from SRS to compute only the necessary detection matrices during data reception, reducing computational load and hardware requirements while maintaining the ability to serve hundreds of antennas and multiple UEs.
3Measurement precision
If conventional detection matrix computation is performed on each radio resource unit, then signal separation for each UE can be achieved, but the process delay increases and cannot meet the typical requirement of Radio Access Network
Solution Approach 1:
The patent merges the channel estimation function with the data detection function by using SRS to acquire channel information that serves multiple purposes: (1) channel estimation for downlink precoding, and (2) preliminary channel knowledge for uplink data detection. This merging allows the BS to use the same channel estimates across multiple resource blocks, reducing the frequency of detection matrix computations and improving processing speed while maintaining signal separation accuracy through selective matrix computation.
Data Source
AI summary
This invention presents methods for signal detection and transmission in MU-MIMO wireless communication systems, for multiple UEs transmit uplink SRS to capture the uplink channel coefficients, and to estimate the frequency offsets and time offsets, for the BS estimating the channel coefficients of multiple UEs with the received SRS, for the BS estimating the frequency offsets and time offsets of multiple UEs with the estimated channel coefficients, for the BS modifying the estimated channel coefficients with the estimated frequency offsets and time offsets, for the BS calculating the detection matrices for the uplink MU-MIMO signal detection with the modified channel coefficients and for the BS calculating the precoding matrices for the downlink MU-MIMO signal transmission.