Digital Signal Processing Reference
In-Depth Information
Fig. 9.11
(
To p
) Non-regularized and (
bottom
) regularized Doppler AR spectrum
We have tested OS-HDR-CFAR on real recorded ground Radar clutter with inges-
tion of synthetic slow targets (Figs.
9.13
,
9.14
).
In Figs.
9.15
,
9.16
, we give ROC curves with Probability of detection versus
probability of false alarm. We observe that OS-HDR-CFAR is better (Pd
=
.
0
8) than
=
.
OS-CFAR/Doppler-filters (Pd
65) for arbitrary fixed Pfa. We could also observe
that Information Geometry approach provides better results than Optimal Transport
Theory approach (based on Wasserstein distance/barycenter: black curve).
To prove that OS-HDR-CFAR is an Ordered-Statistic CFAR, robust against out-
liers, we have compared the case where all targets are ingested every 33 range cells
(CFAR window are limited to 32 range cells), and the case where all targets are
ingested every 7 range cells. We can observe that OS-HDR-CFAR performance are
not altered by targets in the secondary data window (Figs.
9.17
,
9.18
).
0
9.17.2 Robust Space-Time Processing: OS-STAP
We propose in this second applicative part to study robust Space-Time Adaptative
Processing (STAP) based on Median of Sample Covariance Matrix, with advantages
