Title :
On the equivalence of constrained total least squares and structured total least squares
Author :
Lemmerling, Philippe ; Moor, Bart De ; Huffel, Sabine Van
Author_Institution :
ESAT-SISTA, Katholieke Univ., Leuven, Heverlee, Belgium
fDate :
11/1/1996 12:00:00 AM
Abstract :
Several extensions of the total least squares (TLS) method that are able to calculate a structured rank deficient approximation of a data matrix have been developed. The main result of this article is the demonstration of the equivalence of two of these approaches, namely, the constrained total least squares (CTLS) approach and the structured total least squares (STLS) approach. We also present a numerical comparison of both methods
Keywords :
convergence of numerical methods; least squares approximations; matrix algebra; parameter estimation; constrained total least squares; convergence rate; data matrix; linear parameter estimation; numerical comparison; structured rank deficient approximation; structured total least squares; Adaptive signal processing; Artificial intelligence; Geometry; Least squares approximation; Least squares methods; Matrix decomposition; Parameter estimation; Signal processing algorithms; Singular value decomposition; Sparse matrices;
Journal_Title :
Signal Processing, IEEE Transactions on
