Home // ADAPTIVE 2021, The Thirteenth International Conference on Adaptive and Self-Adaptive Systems and Applications // View article

Variable-Regularized Low-Complexity RLS Adaptive Algorithms for Bilinear Forms

Authors:
Ionut-Dorinel Ficiu
Camelia Elisei-Iliescu
Cristian-Lucian Stanciu
Constantin Paleologu

Keywords: adaptive filter; bilinear forms; recursive least-squares; variable regularization; dichotomous coordinate descent

Abstract:
This work in progress targets the identification of bilinear forms using fast converging adaptive algorithms. In this framework, the bilinear term is defined with respect to the impulse responses of a spatiotemporal model. Recently, the recursive least-squares algorithm tailored for bilinear forms (RLS-BF) was introduced in this context, together with its low-complexity version based on dichotomous coordinate descent (DCD) iterations, namely RLS-DCD-BF. Aiming to improve the robustness of the RLS-DCD-BF algorithm in noisy conditions, a variable-regularized version is presented in this paper. Simulation results indicate the appealing performance of this solution and motivate our future works on multilinear forms.

Pages: 9 to 12

Copyright: Copyright (c) IARIA, 2021

Publication date: April 18, 2021

Published in: conference

ISSN: 2308-4146

ISBN: 978-1-61208-848-8

Location: Porto, Portugal

Dates: from April 18, 2021 to April 22, 2021