Channel Estimation Using Spatial Filtering and Virtual Arrays
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current channel estimation techniques, such as the SAGE algorithm, face challenges in accurately estimating a large number of multipath components (MPCs) due to computational complexity and limited feasibility, leading to inaccuracies in channel state information (CSI) for applications like CoMP and massive MIMO systems, especially under varying mobility conditions.
Innovation Solution
The method involves forming virtual multi-antenna arrays by performing measurements at multiple locations and applying spatial filtering to reduce the number of MPCs, combining parameter estimations to produce CSI, and switching between estimation modes based on device velocity, using notch filtering to isolate and suppress strong multipath components, and combining these techniques with increased measurement bandwidth for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional channel estimation techniques (e.g., SAGE algorithm) are used to estimate a large number of multipath components, then channel state information accuracy is improved, but computational complexity increases and estimation feasibility deteriorates
Solution Approach 1:
The patent divides the channel estimation process into two distinct stages: a training phase where channel parameters are estimated using reference signals, and a prediction phase where these parameters are used to predict channel state information. This segmentation separates the computationally intensive parameter estimation from the lighter prediction operations, reducing overall computational complexity while maintaining accuracy.
Solution Approach 2:
The patent performs channel parameter estimation (including multipath component parameters) in advance during a training phase using reference signals before actual data transmission. These pre-estimated parameters are then reused for channel prediction, avoiding the need to repeat complex estimation operations during data transmission and reducing real-time computational burden.
2Measurement precision
If the number of multipath components per tap is increased to improve channel estimation accuracy, then channel state information quality is improved, but the number of parameters to estimate increases, making the problem less feasible
Solution Approach 1:
The patent extracts and estimates only the essential channel parameters (such as multipath component parameters, Doppler frequencies, and angles of arrival/departure) during the training phase. These extracted parameters are then used to predict the full channel state information, avoiding the need to estimate all channel parameters simultaneously and reducing the dimensionality of the estimation problem.
Solution Approach 2:
The patent changes the approach from directly estimating full channel state information to estimating a reduced set of fundamental channel parameters first. By transforming the problem from estimating many correlated channel parameters to estimating fewer independent physical parameters (multipath components, Doppler, angles), the estimation becomes more feasible while maintaining accuracy.
3Productivity
If channel estimation is performed with high accuracy for massive MIMO and CoMP systems, then spectral efficiency is improved, but feedback overhead and processing requirements increase
Solution Approach 1:
The patent creates a simplified copy or model of the channel using estimated parameters (multipath components, Doppler frequencies, angles) during the training phase. This channel model copy is then used to predict future channel states without requiring direct feedback of full channel state information, reducing feedback overhead while maintaining the ability to achieve high spectral efficiency.
Data Source
Figure 1~3
Figure 4~5
Figure 6A
AI summary
Channel estimation where at least two sets of multipath components for a reception range are formed by applying spatial filtering to reference signal information measured by a communication device. Each set of multipath components comprises a number of multipath components that is less than the number of multipath components for the range. Separate parameter estimations are performed on the at least two sets of multipath components. The communication device may be a mobile device, and measurements may be provided by the mobile device in multiple of locations.