Analysis of gradient algorithms for TLS-based adaptive IIR filters
作者： B.E. Dunne , G.A. Williamson
关键词:
摘要: Steepest descent gradient algorithms for unbiased equation error adaptive infinite impulse response (IIR) filtering are analyzed collectively both the total least squares and mixed squares-total framework. These have a monic normalization that allows direct implementation. We show converge to desired filter coefficient vector. achieve convergence result by analyzing stability of equilibrium points demonstrate only solution is locally stable. Additionally, we describe region initialization under which algorithm converges solution. derive results using interlacing relationships between eigenvalues data correlation matrices their respective Schur complements. Finally, illustrate performance these new approaches through simulation.
harvard.edu
acm.org 
sci-hub.se 参考文章(32)
Erkki Oja, Original Contribution: Principal components, minor components, and linear neural networks Neural Networks. ,vol. 5, pp. 927- 935 ,(1992) , 10.1016/S0893-6080(05)80089-9
B.E. Dunne, G.A. Williamson, A locally stable adaptive TLS algorithm asilomar conference on signals, systems and computers. ,vol. 1, pp. 542- 545 ,(1998) , 10.1109/ACSSC.1998.750922
B. WIDROW, M. E. HOFF, Adaptive switching circuits Neurocomputing: foundations of research. pp. 123- 134 ,(1988) , 10.21236/AD0241531
H.C. So, Least mean square algorithm for unbiased impulse response estimation midwest symposium on circuits and systems. ,vol. 2, ,(2002) , 10.1109/MWSCAS.2002.1186823
Stephen Smale, Morris W. Hirsch, Differential Equations, Dynamical Systems, and Linear Algebra ,(1974)
Hideaki Sakai, Kouhei Shimizu, A new adaptive algorithm for minor component analysis Signal Processing. ,vol. 71, pp. 301- 308 ,(1998) , 10.1016/S0165-1684(98)00153-4
Lei Xu, Erkki Oja, Ching Y. Suen, Modified Hebbian learning for curve and surface fitting Neural Networks. ,vol. 5, pp. 441- 457 ,(1992) , 10.1016/0893-6080(92)90006-5
A. Cichocki, R. Unbehauen, Simplified neural networks for solving linear least squares and total least squares problems in real time IEEE Transactions on Neural Networks. ,vol. 5, pp. 910- 923 ,(1994) , 10.1109/72.329687
Shu-An Hu, Ronald L. Smith, The Schur Complement Interlacing Theorem SIAM Journal on Matrix Analysis and Applications. ,vol. 16, pp. 1013- 1023 ,(1995) , 10.1137/S0895479894265022
Sabine Van Huffel, Joos Vandewalle, The Total Least Squares Problem: Computational Aspects and Analysis ,(1987)