Kernel weights for equalizing kernel-wise convergence rates of multikernel adaptive filtering

Kwangjin Jeong, Masahiro Yukawa

研究成果: Article査読

抄録

Multikernel adaptive filtering is an attractive nonlinear approach to online estimation/tracking tasks. Despite its potential advantages over its single-kernel counterpart, a use of inappropriately weighted kernels may result in a negligible performance gain. In this paper, we propose an efficient recursive kernel weighting technique for multikernel adaptive filtering to activate all the kernels. The proposed weights equalize the convergence rates of all the corresponding partial coefficient errors. The proposed weights are implemented via a certain metric design based on the weighting matrix. Numerical examples show, for synthetic and multiple real datasets, that the proposed technique exhibits a better performance than the manually-tuned kernel weights, and that it significantly outperforms the online multiple kernel regression algorithm.

本文言語English
ページ(範囲)927-939
ページ数13
ジャーナルIEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences
1
6
DOI
出版ステータスPublished - 2021 6 1

ASJC Scopus subject areas

  • 信号処理
  • コンピュータ グラフィックスおよびコンピュータ支援設計
  • 電子工学および電気工学
  • 応用数学

フィンガープリント

「Kernel weights for equalizing kernel-wise convergence rates of multikernel adaptive filtering」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル