Title :
Efficient computational schemes for the orthogonal least squares algorithm
Author :
Chng, E.S. ; Chen, S. ; Mulgrew, B.
Author_Institution :
Dept. of Electr. Eng., Edinburgh Univ., UK
fDate :
1/1/1995 12:00:00 AM
Abstract :
The orthogonal least squares (OLS) algorithm is an efficient implementation of the forward selection method for subset model selection. The ability to find good subset parameters with only a linearly increasing computational requirement makes this method attractive for practical implementations. We examine the computational complexity of the algorithm and present a preprocessing method for reducing the computational requirement
Keywords :
computational complexity; least squares approximations; parameter estimation; prediction theory; signal processing; computational complexity; computational requirement reduction; forward selection method; nonlinear predictors; orthogonal least squares algorithm; preprocessing method; subset model selection; subset parameters; Computational complexity; Degradation; Integrated circuit modeling; Least squares methods; Linear regression; Predictive models; Signal processing algorithms; Vectors;
Journal_Title :
Signal Processing, IEEE Transactions on
