|
An Experimental Study of the Eigendecomposition Methods for Blind SIMO System Identification in the Presence of Noise
- Author(s)
-
- Soroush Javidi, (Imperial College)
- Nikolay Gaubitch, (Imperial College)
- Patrick Naylor, (Imperial College)
- Topics
-
- Noise reduction techniques
- Noise and acoustic environments and characteristics
|
Get the paper in PDF format
Acrobat Reader
(version 5 minimum)
is necessary to read this document.
|
Abstract
Subspace methods for blind SIMO system identification have been presented which rely on the null subspace of the data correlation matrix to estimate the impulse response coefficients. It is known that the performance of these algorithms degrades with increasing noise and large system orders. In this paper, we present results of experimental studies that link the performance of the algorithm to the eigenvalues of the correlation matrix. We demonstrate that the eigenvector corresponding to the smallest eigenvalue is not always the best solution in terms of normalised projection misalignment or cross-relation error.
|
