PDAF versus PMHT performance on OTHR data

Author

Colegrove, S.B. ; Davey, S.J. ; Cheung, B.

Author_Institution

DSTO Edinburgh, UK

fYear

2003

fDate

3-5 Sept. 2003

Firstpage

560

Lastpage

565

Abstract

Over the Horizon Radar (OTHR) is the primary Australian wide area surveillance sensor. The radar tracking system currently used for OTHR is based on the Probabilistic Data Association Filter (PDAF). Research and development(R&D)has continued to improve tracking performance. R&D has been extended to include the Probabilistic Multiple Hypothesis Tracker (PMHT). This paper deals with the comparison of the PDAF and the PMHT for tracking targets in experimentally recorded OTHR data. The data has been selected over a range of conditions and every effort has been made to ensure that both trackers have been optimised. The comparison uses a Tracker Assessment Tool (TAT) which measures performance against a total of 15 metrics. This report presents a quantitative performance comparison of the current tracking algorithm and these new tracker developments.

Keywords

interpolation; iterative methods; maximum likelihood estimation; military radar; probability; radar clutter; radar computing; radar signal processing; radar tracking; search radar; sensor fusion; target tracking; expectation maximisation; iterative algorithm; multicomponent nonuniform clutter density; optimal algorithm; over the horizon radar; probabilistic data association filter; probabilistic multiple hypothesis tracker; quadratic interpolation; radar data processing; radar tracking system; wide area surveillance sensor; Australia; Clutter; Data processing; Filters; Force sensors; Radar tracking; Research and development; Springs; Surveillance; Target tracking;

fLanguage

English

Publisher

ieee

Conference_Titel

Radar Conference, 2003. Proceedings of the International

Print_ISBN

0-7803-7870-9

Type

conf

DOI

10.1109/RADAR.2003.1278802

Filename

1278802