vix.ing · top · new · best · stats · spec

Absolute Eigenvalues-Based Covariance Matrix Estimation for a Sparse\n Array

2021/06/07 by Kaushallya Adhikari, Adhikari, Kaushallya
Computer Science · Engineering · #Direction-of-Arrival Estimation Techniques #Speech and Audio Processing #Structural Health Monitoring Techniques

paper · pdf · doi:10.48550/arxiv.2106.03642

Abstract

The ensemble covariance matrix of a wide sense stationary signal spatially\nsampled by a full linear array is positive semi-definite and Toeplitz. However,\nthe direct augmented covariance matrix of an augmentable sparse array is\nToeplitz but not positive semi-definite, resulting in negative eigenvalues that\npose inherent challenges in its applications, including model order estimation\nand source localization. The positive eigenvalues-based covariance matrix for\naugmentable sparse arrays is robust but the matrix is unobtainable when all\nnoise eigenvalues of the direct augmented matrix are negative, which is a\npossible case. To address this problem, we propose a robust covariance matrix\nfor augmentable sparse arrays that leverages both positive and negative noise\neigenvalues. The proposed covariance matrix estimate can be used in conjunction\nwith subspace based algorithms and adaptive beamformers to yield accurate\nsignal direction estimates.\n

Related