Study on joint probability density algorithm in multi-sensor data fusion

Author

Can, Xu ; Zhi, Li

Author_Institution

Co. of Postgrad. Manage., Acad. of Equip., Beijing, China

Volume

1

fYear

2012

fDate

25-27 May 2012

Firstpage

32

Lastpage

37

Abstract

Joint probability density algorithm (JPDA) is a novel algorithm in multi-sensor data fusion and it provides a new approach for target localization. Generally, more sensors could make higher precision of localization when JPDA is adopted, but there is no full theoretical support so far. According to the region of interesting (ROI) generated by JPDA in Cartesian coordinates, the information entropy of JPDA is analyzed. The expression of information entropy of multi-sensor which is the theoretical basis of JPDA is deduced. The result indicates that no matter how many sensors there are, the entropy is determined only by the determinant of covariance matrix, more sensors make information entropy lower which is the reason why localization precision is higher. Simulation results verify our analysis.

Keywords

covariance matrices; entropy; probability; sensor fusion; Cartesian coordinates; JPDA; ROI; covariance matrix; information entropy; joint probability density algorithm; multisensor data fusion; region of interesting; target localization; Coordinate measuring machines; Covariance matrix; Entropy; Information entropy; Joints; Probability density function; Sensors; informatin entropy; joint probabiltiy density; multi-sensor data fusion;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Science and Automation Engineering (CSAE), 2012 IEEE International Conference on

Conference_Location

Zhangjiajie

Print_ISBN

978-1-4673-0088-9

Type

conf

DOI

10.1109/CSAE.2012.6272542

Filename

6272542