Statistical performance comparison of high-resolution DOA estimation methods

Author

Huang, Jianguo ; Chen, Jianfeng

Author_Institution

Northwestern Polytech. Univ., Xian, China

Volume

1

fYear

1996

fDate

14-18 Oct 1996

Firstpage

505

Abstract

A wide comparison of statistical performance among high-resolution methods of direction estimation (MODE) is presented. The development of MODE is first reviewed. In order to investigate their statistical performance clearly, more than 30 kinds of methods are studied and divided into 5 groups. Ten typical approaches are evaluated in resolution and precision respectively. Based on a united simulation model, their performance are compared by using Monte-Carlo statistical analysis. Furthermore, several criteria for the application of MODE are presented to judge their practicality. The theoretical study shows the performance of weighted subspace fitting (WSF) possesses, and indicates that MUSIC, Mini-Norm, TLS-ESPRIT are the most promising methods in practice

Keywords

Monte Carlo methods; array signal processing; direction-of-arrival estimation; signal resolution; statistical analysis; MUSIC; Mini-Norm; Monte-Carlo statistical analysis; TLS-ESPRIT; direction estimation methods; high resolution DOA estimation methods; multisensor array; precision; statistical performance comparison; united simulation model; weighted subspace fitting process; Analytical models; Array signal processing; Direction of arrival estimation; Educational institutions; Maximum likelihood estimation; Multiple signal classification; Noise robustness; Signal resolution; Statistical analysis; Turning;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing, 1996., 3rd International Conference on

Conference_Location

Beijing

Print_ISBN

0-7803-2912-0

Type

conf

DOI

10.1109/ICSIGP.1996.567313

Filename

567313