Title

Efficient Implementations Of Complex Block Conjugate Gradient Lms With Individual Adaptation

Abstract

Recent advances in signal processing hardware has made possible the implementation of sophisticated algorithms. The Complex Block Conjugate Least Mean Square algorithm with Individual adaptation (CBCI-LMS) has recently been proposed and applied in adaptive filtering applications. This technique involves matrix inversion, which is computationally intensive. To improve the feasibility of the algorithm in practice, especially for high-order adaptive systems, this paper proposes two efficient implementations of the CBCI-LMS algorithm. The first efficient implementation applies an iterative matrix inversion lemma without any sacrifice in convergence speed and accuracy. The second method further simplifies the algorithm by using a diagonal matrix for approximation at the first iteration, at the expense of a few additional iterations to achieve convergence. Both of the simplified approaches effectively reduce the computational complexities of the CBCI-LMS algorithm. © 2013 IEEE.

Publication Date

12-1-2013

Publication Title

Midwest Symposium on Circuits and Systems

Number of Pages

1403-1406

Document Type

Article; Proceedings Paper

Personal Identifier

scopus

DOI Link

https://doi.org/10.1109/MWSCAS.2013.6674919

Socpus ID

84893194100 (Scopus)

Source API URL

https://api.elsevier.com/content/abstract/scopus_id/84893194100

This document is currently not available here.

Share

COinS