Aerospace and Electronic Systems Magazine September 2017 - 54

Focus Before Detection: Part I

Figure 5.

Integration results without noise interference. (a) MTD. (b) KT+MTD. (c) HT. (d) RFT.

MTD (KT+MTD) [32]; HT [23]; and RFT as in (2). The MTD accomplishes coherent integration via the Doppler filter bank, where
the ARC and ADC problems are not considered. The KT+MTD
method accomplishes ARC compensation and coherent integration
via KT and MTD, respectively. The HT and the RFT project the
echoes' energy into the parameter space without and with phase
compensation, respectively. Then, for noise-free and real noise interference, the outputs of MTD, KT+MTD, HT, and RFT are given
as Figures 5 and 6, respectively, from which some conclusions can
be drawn as follows:
C

Because of the uncompensated ARC effect, the MTD outputs are submerged in the system noise as Figure 6a, which
is unsuitable for detection of high-speed targets with the
ARC effect.
The KT method can compensate for the linear ARC, i.e.,
range walk. Therefore, the peaks of T1 and T3 are sharp
(Figure 5b) and higher than the noise level (Figure 6b).
However, T2 is missed, as shown in Figure 6b, because of
the uncompensated Doppler ambiguity.
The HT method can also compensate for the linear ARC and
generate the three peaks in Figure 5c. However, because the
phase modulation caused by RM is not compensated, the
output SNR of HT is unacceptable in the real noise background and targets are nearly missed (Figure 6c).

The RFT method in (2) can compensate for both linear ARC
and phase modulation to generate sharp peaks (Figure 5d).
Because of the high SNR gain, these three peaks can be detected (Figure 6d). Besides, the target positions and motion
parameters can be estimated according to the peak locations.
Doppler ambiguity has also been resolved by RFT for highspeed target T2, because the searched velocity of RFT can be
larger than the blind speed with envelope and phase compensations to overcome the Doppler ambiguity.

MANEUVERING TARGET DETECTION VIA GRFT
For the detection of a maneuvering target with acceleration, the
second-order GRFT is proposed in (3) to compensate for phase
modulation and envelope shift caused by the acceleration in a long
TOT. Similar to RFT, the proposed second-order GRFT is also the
LRT [39]-[41] for detection of maneuvering targets, i.e., the statistically optimal detector in an AGWN background. In addition,
based on the system parameters used in the previous subsection,
the results of the second-order GRFT are given for an accelerated
target with parameters of 220 km, 80 m/s, and 2.0 m/s2 on different
projection planes, as shown in Figures 7a-7c, respectively. With
the sharp peak generated by the proposed second-order GRFT
in different planes as in Figures 7a-7c, the target detection and
parameter estimation can be easily accomplished by the CFAR

IEEE A&E SYSTEMS MAGAZINE

SEPTEMBER 2017

Table of Contents for the Digital Edition of Aerospace and Electronic Systems Magazine September 2017