RFS MCMC Predetection Fusion Applied to Multistatic Sonar Data

Author

Georgescu, Ramona ; Willett, Peter

Author_Institution

Dept. of Electr. & Comput. Eng., Univ. of Connecticut, Storrs, CT, USA

Volume

48

Issue

4

fYear

2012

fDate

10/1/2012 12:00:00 AM

Firstpage

2894

Lastpage

2907

Abstract

Predetection fusion can be indispensable for multisensor/multitarget tracking using large networks of low quality sensors. Previously we derived both the "optimal" generalized likelihood ratio test (GLRT) and a more practicable contact-sifting variant. Unfortunately, the gaps between the two in terms both of computation time and performance are not inconsiderable. In this paper we propose an approach, based on random finite sets (RFS) and implemented by Markov chain Monte Carlo (MCMC) simulation, that offers a good balance between run time and metrics for the tracking results.

Keywords

Markov processes; Monte Carlo methods; radar signal processing; radar tracking; sensor fusion; sonar; target tracking; Markov chain Monte Carlo simulation; RFS MCMC predetection fusion; contact-sifting variant; low quality sensors; multisensor/multitarget tracking; multistatic sonar data; optimal generalized likelihood ratio test; random finite sets; run time; Measurement uncertainty; Monte Carlo methods; Receivers; Sensor fusion; Sonar; Target tracking;

fLanguage

English

Journal_Title

Aerospace and Electronic Systems, IEEE Transactions on

Publisher

ieee

ISSN

0018-9251

Type

jour

DOI

10.1109/TAES.2012.6324668

Filename

6324668