Animportant issue inradar space-time adaptive processing (STAP) istheselection oftraining data forestimation ofthe unknowndisturbance covariance matrix. Inthis paper, we address thetraining data selection issue andanalyze theperformance ofnormalized adaptive matched filter (NAMF)in heterogeneous non-Gaussian radar clutter scenarios. Simulations oftheprobability ofdetection (PD)versus signal-tointerference-plus-noise ratio (SINR) fortheNAMF areconducted using Mountain Topdata, whichexhibit non-Gaussian statistics. Toselect thetraining data representative oftheinterference, weapply adaptive beamforming andshowthat PD versus SINRforNAMF isrobust withrespect toawiderange ofvariation inthebeamformer output. Tomitigate heterogeneous clutter, theself-censoring reiterative fast maximum likelihood (SCRFMIL) algorithm isemployed toregularize the eigen spectrum underlying theunknown disturbance covariancematrix. We demonstrate that theNAMF detection performance canbesignificantly improved withtheapplication ofSCRFML.
[1]
William L. Melvin,et al.
Screening among Multivariate Normal Data
,
1999
.
[2]
R. Nitzberg.
An effect of range-heterogeneous clutter on adaptive Doppler filters
,
1990
.
[3]
William L. Melvin,et al.
Improving practical space-time adaptive radar
,
1997,
Proceedings of the 1997 IEEE National Radar Conference.
[4]
Muralidhar Rangaswamy,et al.
Statistical analysis of the nonhomogeneity detector for non-Gaussian interference backgrounds
,
2005,
IEEE Transactions on Signal Processing.
[5]
B. Himed,et al.
Improved detection of close proximity targets using two-step NHD
,
2000,
Record of the IEEE 2000 International Radar Conference [Cat. No. 00CH37037].
[6]
William L. Melvin,et al.
Space-time adaptive radar performance in heterogeneous clutter
,
2000,
IEEE Trans. Aerosp. Electron. Syst..