IEEE Aerospace and Electronic Systems Magazine - October 2020 - 48

Incoherent Imaging at Microwave and Millimeter-Wave Frequencies Using Noise Transmitters
splitter was connected to one transmitter, whereas the
other output was connected to the second transmitter
through a 7.6 m cable. This additional delay ensured that
the transmitted signals were temporally uncorrelated
when the signals were incident on the scene. An additional
9 dB gain amplifier with a 1 dB compression point of
19 dBm was used to overcome the losses from the splitter
and the 7.6 m cable. The three transmit noise signals had a
maximum power of 0 dBm and were connected to 20 dBi
standard-gain horn antennas.
The receiver consisted of two wideband horn antennas
with approximately 10 dBi gain at 5.85 GHz. The synthesized array was an inverse T-array with 39 total element
locations, synthesized by moving the two receiving antennas
on a metal rack; no redundant baselines were collected. By
moving one in the horizontal and the other in the vertical
direction, the T-array was synthesized with a maximum
spacing of 15 in the horizontal axis and 8 in the vertical
axis, in 0.5 increments. The narrow baselines of less than
2 were omitted due to the physical dimensions of the horn
antennas that were used. The two received signals were
amplified with 20 dB low-noise amplifiers (LNAs) and
downconverted to baseband using quadrature mixers with a

Figure 7.
(a) Experimental configuration for the 5.85 GHz experimental
measurements. (b) Block diagram of the experimental imaging
system [17].

5.85 GHz local oscillator (LO). The four signals (I and Q for
each of the two antennas) were then captured using a
20 GHz Keysight MSOX92004A oscilloscope in high resolution mode. The collected signals were processed offline in
MATLAB. The received signals were first low-pass filtered
to a bandwidth of 25 MHz, the dc bias was removed by subtracting the average, and cross-correlation was applied to
each antenna pair. The total integration time was 10 ms. The
utilized bandwidth and integration time are both significantly lower than those typically required in passive interferometric imaging, where bandwidths up to multiple GHz and
integration times beyond 1 s are not uncommon [17]. Thus,
the use of active transmission enables faster imaging with
far less costly hardware than passive systems.
Experimental image reconstruction of two metal
spheres, one 18 cm in radius and one 12 cm in radius, is
shown in Figure 8(a). The two spheres were spaced 60 cm
apart on a pedestal in a 7.6 m semianechoic arch range,
with microwave absorber placed behind the spheres. Their
azimuth angle was Æ0:05 rad. The semianechoic arch
range was smaller than the Rayleigh distance of the
receiving aperture, and the targets were not located in the
far-field, however, because they are located near broadside
the phase errors are minimal compared to the far-field
approximation [6], [28]. Figure 8(b) shows the reconstructed spheres, after applying Gaussian smoothing on
the reconstructed image. The left response clearly indicates the larger sphere and the lower intensity response on
the right indicates the smaller sphere.

Figure 8.
(a) Optical image of the two reflecting spheres. (b) 5.85 GHz
image reconstruction of the two spheres [17].

IEEE A&E SYSTEMS MAGAZINE

OCTOBER 2020

IEEE Aerospace and Electronic Systems Magazine - October 2020

Table of Contents for the Digital Edition of IEEE Aerospace and Electronic Systems Magazine - October 2020