A modified exact maximum likelihood registration algorithm

Author

Yunlong, Dong ; You, He ; Guohong, Wang

Author_Institution

Res. Inst. of Inf. Fusion, Naval Aeronaut. Eng. Inst., Yantai, China

fYear

2004

fDate

1-4 Nov. 2004

Firstpage

85

Lastpage

88

Abstract

The registration problem is a prerequisite process for radar networking systems to estimate and correct systematic errors accurately. An exact maximum likelihood (EML) registration algorithm was presented in the literature (Zhou Yifeng et al, IEEE Trans. Sigpro., vol.45(6), p.1560-1572, 1997), which incorporates the effects of measurement noise. Although this algorithm is superior to some classical registration algorithms, we find its likelihood function is not exact. A novel modified maximum likelihood (MEML) registration algorithm is presented, which is superior to the EML registration algorithm, with an exact likelihood function. Finally, simulated data are used to evaluate the performance of the two algorithms.

Keywords

maximum likelihood estimation; measurement uncertainty; radar theory; sensor fusion; EML registration algorithm; MEML; exact likelihood function; exact maximum likelihood registration algorithm; measurement noise effects; multisensor fusion; radar networking systems; systematic error correction; Azimuth; Covariance matrix; Density functional theory; Error correction; Helium; Least squares approximation; Maximum likelihood estimation; Noise measurement; Radar; Wide area networks;

fLanguage

English

Publisher

ieee

Conference_Titel

Computational Electromagnetics and Its Applications, 2004. Proceedings. ICCEA 2004. 2004 3rd International Conference on

Print_ISBN

0-7803-8562-4

Type

conf

DOI

10.1109/ICCEA.2004.1459295

Filename

1459295