DocumentCode :
3110862
Title :
A robust Doppler classification technique based on hidden Markov models
Author :
Jahangir, M. ; Ponting, K.M. ; Loghlen, J. W O
Author_Institution :
QinetiQ Malvern, UK
fYear :
2002
fDate :
15-17 Oct. 2002
Firstpage :
162
Lastpage :
166
Abstract :
A classification algorithm is presented that uses hidden Markov models (HMM) to carry out recognition between three classes of targets: personnel, tracked vehicles and wheeled vehicles. It exploits the time varying nature of radar Doppler data in a manner similar to techniques used in speech recognition, albeit with a modified topology, to distinguish targets with different Doppler characteristics. The algorithm was trained and tested on real radar signatures of multiple examples of moving targets from each class and the performance was shown to be invariant to target speed and orientation and was able to generalise with respect to variants within a class.
Keywords :
Doppler effect; hidden Markov models; radar target recognition; signal classification; Doppler characteristics; HMM algorithm design; classification algorithm; hidden Markov models; moving target indication radar; moving targets; optimum model topology; personnel; real radar signatures; robust Doppler classification; target orientation; target recognition; target speed; time varying radar Doppler data; tracked vehicles; wheeled vehicles; Classification algorithms; Doppler radar; Hidden Markov models; Personnel; Radar tracking; Robustness; Speech recognition; Target recognition; Target tracking; Vehicles;
fLanguage :
English
Publisher :
iet
Conference_Titel :
RADAR 2002
Conference_Location :
Edinburgh, UK
ISSN :
0537-9989
Print_ISBN :
0-85296-750-0
Type :
conf
DOI :
10.1109/RADAR.2002.1174674
Filename :
1174674
Link To Document :
بازگشت