Pipelined Cordic based QRD-RLS adaptive filtering using matrix lookahead

Author

Jun Ma ; Deprettere, Ed F. ; Parhi, Keshab K.

Author_Institution

Dept. of Electr. & Comput. Eng., Minnesota Univ., Minneapolis, MN, USA

fYear

1997

fDate

3-5 Nov 1997

Firstpage

131

Lastpage

140

Abstract

In this paper, the matrix lookahead transformation is developed to achieve fine-grain pipelining for Cordic based QRD-RLS adaptive filtering algorithm. Various implementation styles are proposed. They include pipelining, block processing, and incremental block processing. The proposed architectures can operate at arbitrarily high sample rates, and consist of only Givens and a few Gaussian rotations which can be mapped onto a Cordic based Jacobi specific dataflow processor

Keywords

adaptive filters; adaptive signal processing; least squares approximations; pipeline processing; Cordic based; Gaussian rotations; Givens; Jacobi specific dataflow processor; QRD-RLS; adaptive filtering; block processing; fine-grain pipelining; incremental block processing; pipelining; Adaptive filters; Adaptive signal processing; Arithmetic; Array signal processing; Filtering algorithms; Jacobian matrices; Least squares methods; Pipeline processing; Signal processing algorithms; USA Councils;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing Systems, 1997. SIPS 97 - Design and Implementation., 1997 IEEE Workshop on

Conference_Location

Leicester

ISSN

1520-6130

Print_ISBN

0-7803-3806-5

Type

conf

DOI

10.1109/SIPS.1997.626110

Filename

626110