RT Journal Article
T1 Combined nonlinear filtering architectures involving sparse functional link adaptive filters
A1 Comminielle, Danilo
A1 Scarpiniti, Michele
A1 Azpicueta Ruiz, Luis Antonio
A1 Arenas García, Jerónimo
A1 Uncini, Aurelio
AB Sparsity phenomena in learning processes have been extensively studied, since their detection allows to derive suited regularized optimization algorithms capable of improving the overall learning performance. In this paper, we investigate the sparsity behavior that may occur in nonlinear adaptive filtering problems and how to leverage it and develop enhanced algorithms. In particular, we focus on a particular class of linear-in-the-parameters nonlinear adaptive filters, whose nonlinear transformation is based on a functional link expansion. The analysis of the sparsity in functional links leads us to derive a family of adaptive combined filtering architectures that is capable of exploiting any sparseness degree in the nonlinear filtering. We propose two different filtering schemes based on a new block-based combined approach, well suited for sparse adaptive algorithms. Moreover, a hierarchical architecture is also proposed that generalizes the different combined schemes and does not need any a priori information about the nature of the nonlinearity to be modeled. Experimental results prove the effectiveness of the proposed combined architectures in exploiting any sparseness degree and improving the modeling performance in nonlinear system identification problems.
PB Elsevier
SN 0165-1684
YR 2017
FD 2017-06
LK https://hdl.handle.net/10016/33885
UL https://hdl.handle.net/10016/33885
LA eng
NO The work of M. Scarpiniti and A. Uncini is partially supported by the Italian National Project "GAUChO - A Green Adaptive Fog Computing and Networking Architecture", under grant number 2015YPXH4W. The work of L. A. Azpicueta-Ruiz is partially supported by Comunidad de Madrid under grant 'CASI-CAM-CM' (id. S2013/ICE-2845), by the Spanish Ministry of Economy and Competitiveness (under grant DAMA (TIN2015-70308-REDT) and grant TEC2014-52289-R), and by the European Union. The work of J. Arenas-García has been partly funded by MINECO project TEC2014-52289-R, and by Comunidad de Madrid project PRICAM S2013/ICE-2933.
DS e-Archivo
RD 27 jul. 2024