• SAR
• ISAR
• signal processing

This thesis reports on research into the eld of Synthetic Aperture Radar
(SAR) and Inverse Synthetic Aperture Radar (ISAR) signal processing. The
contributions of this thesis may be divided into two following parts:
 A new bistatic 3D near eld circular SAR imaging algorithm was devel-
oped. High resolution radar imaging is typically obtained by combining
wide bandwidth signals and synthetic aperture processing. High range
resolution is obtained by using modulated signals whereas high cross
range resolution is achieved by coherently processing the target echoes
at dierent aspect angles of the target. Anyway, theoretical results have
shown that when the aspect angle whereby the target is observed is suf-
ciently wide, high resolution target images can be obtained by using
continuous wave (CW) radars [2], therefore allowing to reduce hardware
costs. In a similar way, three dimensional radar imaging can be per-
formed by coherently processing the backscattered eld as a function of
two rotation angles about two orthogonal axes [3].Three dimensional tar-
get radar imaging can be eciently obtained by means of a 3D Fourier
Transform, when the far-eld (planar wave) approximation holds. Oth-
erwise, the wavefront curvature has to be accounted for. For this reason,
a new algorithm based on a near eld spherical wave illumination that
takes into account the wavefront curvature by adopting a planar piece-
wise approximation was designed. This means that the wavefront is as-
sumed to be locally planar around a given point on the target. The oper-
ator that the algorithm uses for the focusing procedure is a space variant
focusing function which aims at compensating the propagation losses and
the wavefront curvature. The algorithm has been developed under the
Microwave Electronic Imaging Security and Safety Access (MELISSA)
project. The system MELISSA is a body scanner whose purpose is the
detection of concealed objects. The added value of the system is the
capability to provide an electromagnetic image of the concealed objects.
The author would like to thank all people that worked at the project, all
LabRass colleagues, all people who designed and acquired real data, all people that permitted the drafting of the rst part of this thesis. The
developed algorithm was presented in the chapter 1. The goal of this
work was the system design concerning the imaging point of view, by
simulating and therefore predicting the system performance by means of
the developed algorithm. In the chapter 2 was shown how the design was
achieved. Finally, in the chapter 3, the results on real data measured in
anechoic chamber with a system with characteristics very close to the
nal system prototype MELISSA, was presented.
 A new way of ISAR processing has been dened, by applying the tradi-
tional ISAR processing to data acquired from passive radars. Purpose of
the ISAR processing is to extract an electromagnetic bi-dimensional im-
age of the target in order to determine the main geometric features of the
target, allowing (when possible) recognition and classication. Passive
radars are able to detect and track targets by exploiting illuminators of
opportunity (IOs). In this work of thesis, it will be proven that the same
concept can be extended to allow for Passive Inverse Synthetic Aperture
Radar (P-ISAR) imaging. A suitable signal processing is detailed that
is able to form P-ISAR images starting from range-Doppler maps, which
represent the output of a passive radar signal processing. Multiple chan-
nels Digital Video Broadcasting - Terrestrial (DVB-T) signals are used to
demonstrate the concept as they provide enough range resolution to form
meaningful ISAR images. The problem of grating lobes, generated by
DVB-T signal, is also addressed and solved by proposing an innovative
P-ISAR technique. The second part of this thesis has been developed un-
der the Array Passive ISAR adaptive processing (APIS) project. APIS is
dened as a multichannel, bi-static single receiver for array passive radar,
capable of detecting targets and generating ISAR images of the detected
targets for classication purposes. The author would like to thank all
people that worked at the project, all LabRass colleagues, all people who
designed, built the prototype and acquired real data, all people that per-
mitted the drafting of the second part of this thesis. In the chapter 4, the
basics on Passive Bistatic Radar (PBR) was brie
y recalled, the P-ISAR
processor was detailed and the new algorithm per the Grating Lobes
Cancellation was presented. In the chapter 5, some numerical results
on simulated data was shown, in order to demonstrate the potentiality
of the P-ISAR, for the imaging and classication purpose. In fact, by
using more than three adjacent channels and by observing the signal for
a long time, ner range and cross-range resolutions, respectively, could
be achieved. Finally, the obtained results on real data was discussed in
the chapter 6.