TY - JOUR
T1 - Adaptive nonlinear estimation based on parallel projection along Affine subspaces in reproducing kernel Hilbert space
AU - Takizawa, Masa Aki
AU - Yukawa, Masahiro
N1 - Funding Information:
This work was supported by KDDI Foundation and JSPS Grants-in-Aid (24760292). This work was partially presented at EUSIPCO2012 [1] and ICASSP 2013 [2].
Publisher Copyright:
© 2015 IEEE.
PY - 2015/8/15
Y1 - 2015/8/15
N2 - We propose a novel algorithm using a reproducing kernel for adaptive nonlinear estimation. The proposed algorithm is based on three ideas: projection-along-subspace, selective update, and parallel projection. The projection-along-subspace yields excellent performances with small dictionary sizes. The selective update effectively reduces the complexity without any serious degradation of performance. The parallel projection leads to fast convergence/tracking accompanied by noise robustness. A convergence analysis in the non-selective-update case is presented by using the adaptive projected subgradient method. Simulation results exemplify the benefits from the three ideas as well as showing the advantages over the state-of-the-art algorithms. The proposed algorithm bridges the quantized kernel least mean square algorithm of Chen and the sparse sequential algorithm of Dodd et al.
AB - We propose a novel algorithm using a reproducing kernel for adaptive nonlinear estimation. The proposed algorithm is based on three ideas: projection-along-subspace, selective update, and parallel projection. The projection-along-subspace yields excellent performances with small dictionary sizes. The selective update effectively reduces the complexity without any serious degradation of performance. The parallel projection leads to fast convergence/tracking accompanied by noise robustness. A convergence analysis in the non-selective-update case is presented by using the adaptive projected subgradient method. Simulation results exemplify the benefits from the three ideas as well as showing the advantages over the state-of-the-art algorithms. The proposed algorithm bridges the quantized kernel least mean square algorithm of Chen and the sparse sequential algorithm of Dodd et al.
KW - Convex projection
KW - kernel adaptive filtering
KW - reproducing kernel Hilbert space
UR - http://www.scopus.com/inward/record.url?scp=84960077626&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84960077626&partnerID=8YFLogxK
U2 - 10.1109/TSP.2015.2437835
DO - 10.1109/TSP.2015.2437835
M3 - Article
AN - SCOPUS:84960077626
SN - 1053-587X
VL - 63
SP - 4257
EP - 4269
JO - IEEE Transactions on Acoustics, Speech, and Signal Processing
JF - IEEE Transactions on Acoustics, Speech, and Signal Processing
IS - 16
M1 - 7112637
ER -