High-Resolution Azimuth Detection of Targets in HFSWRs Under Strong Interference

1.

High-Resolution Azimuth
Detection of Targets in HFSWRs
Under Strong Interference
Dragan Golubović, Miljko Erić and Nenad Vukmirović

2.

slide 1/15
Introduction
• Maritime surveillance of vessels at long distances is up-todate research topic!
• The focus –>> Accurate azimuth detection of targets
presents a very important challenge because HFSWR's
performance directly depends on the detection quality! Main
problem –>> strong external interference!
vessel location estimation…

3.

slide 2/15
Introduction
• The proposed method –>> The azimuth estimation is based on
the eigenvalue decomposition of the signal covariance matrix.
In this case, the dimensionality of the covariance matrix is
smaller compared to the RD-HR map covariance matrix, but it
also requires the knowledge of the dimensionality of the
signal subspace (parameter KA of the MUSIC-based
algorithm).
• In real situations this parameter is often set to 1 or 2, and in
most cases that is enough for accurate azimuth estimation!

4.

slide 3/15
Introduction
• When the strong interference is present, the parameter KA
has to be greater, because some peaks of the criterion
function cannot be detected!
• Conventional algorithms such as AIC MDL tend to
overestimate the value of KA. Thus, this parameter can be
determined using the proposed method or can be set to any
greater value, which can be a nonoptimal solution, because
false alarm probability can be increased in that case.

5.

slide 4/15
HIGH-RESOLUTION AZIMUTH
DETECTION OF TARGETS
Azimuth estimation of an arbitrarily chosen
target from the RD-HR map
• While creating a RD-HR
map, phase data were
lost, which was not the
case while creating an
RD-FFT map.
• Since these phases are
necessary in the
procedure for azimuth
estimation, it is
necessary to return to
the previously formed
FFT matrix...

6.

slide 5/15
HIGH-RESOLUTION AZIMUTH
DETECTION OF TARGETS
The selection of the appropriate
column of the FFT matrix for the
selected target

7.

slide 6/15
METHOD FOR IMPROVED
AZIMUTH ESTIMATION OF TARGETS IN THE
PRESENCE OF STRONG INTERFERENCE
• Interference exists in the third domain (spatial domain), where external
interference cannot be detected at all on the RD-HR map (detections
are distributed differently on it).
• The key to this new method for enhanced azimuth estimation is the
appropriate parameter