Regularized Tapered Covariance Estimation Under Limited Samples
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In wireless communication systems, especially in future networks with multiple antennas, interference from intra-cell and inter-cell sources significantly degrades detection performance, and existing methods for interference reduction are inadequate, particularly when the number of antennas exceeds the number of available samples for covariance matrix estimation.
Innovation Solution
A method for estimating the interference-plus-noise covariance matrix using a regularized tapered sample covariance matrix (RTSCM) with optimized shrinkage parameters and tapering matrices, chosen through leave-one-out cross-validation, to improve estimation accuracy and robustness against moving interferers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the number of antennas is increased to improve detection performance, then the ability to reject interference is improved, but the accuracy of covariance matrix estimation deteriorates when the number of samples is limited
Solution Approach 1:
The patent applies parameter changes by modifying the covariance matrix estimation process through regularization (adding a scaled identity matrix) and tapering (applying a spectral tapering function). These parameter transformations allow the system to maintain stable estimates even when the number of antennas exceeds the number of samples, thus resolving the contradiction between improved interference rejection and maintained estimation accuracy
Solution Approach 2:
The patent creates a composite estimation approach by combining the sample covariance matrix with a scaled identity matrix through regularization, and further combining this with spectral tapering. This composite structure ensures the estimator remains well-conditioned and accurate even in high-dimensional scenarios with limited samples
2Ease of manufacture
If conventional covariance matrix estimation methods are used, then the implementation is simple, but the estimation accuracy deteriorates when the number of antennas exceeds the number of samples
Solution Approach 1:
The patent modifies the standard covariance matrix computation by introducing regularization parameter α and spectral tapering function g(λ). These parameter changes transform the estimation process to maintain accuracy in high-dimensional settings while preserving computational tractability through efficient matrix operations
Solution Approach 2:
The patent performs preliminary conditioning of the covariance matrix estimate by applying regularization before using it in detection algorithms. This preliminary action ensures the matrix is well-conditioned and accurate even when antennas exceed samples, preventing numerical instability downstream
Data Source
AI summary
Apparatuses and methods in a communication system are disclosed. A signal is received. The signal comprises demodulation reference information. The apparatus compute estimates of the channel and of interference-plus-noise vectors. The interference-plus-noise vectors are used as input for choosing a tapering matrix and for determining shrinkage target matrix and parameter. An interference plus noise covariance matrix estimator is computed and used in an equalizer, which has as input an estimate of the interference-plus-noise covariance matrix.