An iterative blind cyclostationary beamforming algorithm

Author

Du, K.-L. ; Swamy, M.N.S.

Author_Institution

Dept. of Electr. & Comput. Eng., Concordia Univ., Montreal, Que., Canada

Volume

1

fYear

2002

fDate

2002

Firstpage

145

Lastpage

148

Abstract

The cross-correlation neural network proposed in Diamantaras and Kung (1994) is an efficient iterative method for singular value decomposition. In this paper, we propose an iterative blind cyclostationary beamforming algorithm, which is inspired by the cross-correlation neural model. It can be used to extract signals with cyclostationarity. The new algorithm is a gradient decent-based method. It is fast, simple, and easy to implement. Simulation shows that it can provide good performance as long as the learning rate is suitably selected

Keywords

antenna theory; array signal processing; iterative methods; linear antenna arrays; neural nets; principal component analysis; cross-correlation neural model; cyclostationary signals; gradient decent-based method; iterative blind cyclostationary beamforming algorithm; learning rate; signals extraction; Antenna arrays; Array signal processing; Covariance matrix; Directive antennas; Eigenvalues and eigenfunctions; Iterative algorithms; Neural networks; Principal component analysis; Signal processing algorithms; Singular value decomposition;

fLanguage

English

Publisher

ieee

Conference_Titel

Communications, 2002. ICC 2002. IEEE International Conference on

Conference_Location

New York, NY

Print_ISBN

0-7803-7400-2

Type

conf

DOI

10.1109/ICC.2002.996834

Filename

996834