Efficient total least squares method for system modeling using minor component analysis

Y.N. Rao1, J.C. Principe1
1Computational NeuroEngineering Laboratory, University of Florida, Gainesville, FL, USA

Tóm tắt

We present two algorithms to solve the total least-squares (TLS) problem. The algorithms are on-line with O(N/sup 2/) and O(N) complexity. The convergence of the algorithms is significantly faster than the traditional methods. A mathematical analysis of convergence is also provided along with simulations to substantiate the claims. We also apply the TLS algorithms for FIR system identification with known model order in the presence of noise.

Từ khóa

#Least squares methods #Modeling #Signal processing algorithms #Parameter estimation #Algorithm design and analysis #Convergence #Vectors #Analytical models #Neural engineering #Laboratories

Tài liệu tham khảo

haykin, 1986, Adaptive Filter Theory rao, 2000, A Fast On-line Algorithm for PCA and its Convergence Characteristics, Proc IEEE Workshop Neural Networks Signal Process, 299 rao, 2000, Algorithms For Eigendecomposition and Time Series Segmentation rao, 0, The CNEL rule: A Novel On-line Algorithm for Principal Component Analysis 10.1109/TAC.1977.1101561 10.1007/978-3-642-75894-2 10.1016/0893-6080(95)91644-9 10.1109/78.700968 diamantaras, 1996, Principal Component Neural Networks, Theory and Applications moon, 1999, Mathematical Methods and Algoritbms for Signal Processing golub, 1989, Matrix Cemputatlens 10.1109/78.275601 10.1109/72.839022 10.1109/78.348123 10.1016/0893-6080(92)90006-5 luo, 1997, A Minor Subspace Analysis Algorithm, IEEE Trans Neural Networks, 8 gao, 1994, A Constrained Anti-Hebbian Learning Algorithm for Total Least-Squares Estimation with Applications to Adaptive FIR and IIR Filtering, IEEE Trans Circuits and Systems-II Analog and Digital Signal Processing, 41 deprettere, 1988, Applications and Architectures, SVD And Signal Processing - Algorithms 10.1109/82.544031