This paper introduces a variable regularization method for the fast affine projection algorithm (VR-FAP). It is inspired by a recently introduced technique for variable regularization of the classical, affine projection algorithm (VR-APA). In both algorithms, the regularization parameter varies as a function of the excitation, measurement noise, and residual error energies. Because of the dependence on the last parameter, VR-APA and VR-FAP demonstrate the desirable property of fast convergence (via a small regularization value) when the convergence is poor and deep convergence/immunity to measurement noise (via a large regularization value) when the convergence is good. While the regularization parameter of APA is explicitly available for on-line modification, FAP's regularization is only set at initialization. To overcome this problem we use noise-injection with the noise-power proportional to the variable regularization parameter. As with their fixed regularization versions, VR-FAP is considerably less complex than VR-APA and simulations verify that they have the very similar convergence properties
S. L. Grant et al., "Variable Regularized Fast Affine Projections," Proceedings of the 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07, Institute of Electrical and Electronics Engineers (IEEE), Apr 2007.
The definitive version is available at https://doi.org/10.1109/ICASSP.2007.366623
2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07
Electrical and Computer Engineering
Keywords and Phrases
Adaptive Filters; Filtering Theory
Article - Conference proceedings
© 2007 Institute of Electrical and Electronics Engineers (IEEE), All rights reserved.