الفهرس 
Spectral Estimation TechniquesOnly 14 pages are availabe for public view
 |  | from | 198 |  |  |
| | | from | 198 | | |
Abstract
This dissertation addresses the problem of detecting multiple wide band sources and estimating
their angles of arrival and the frequency spectrum based on the signal received by a sensor array
in the presence of noise. The sources are considered as wide-band incoherent/coherent signals.
Most of super-resolution Direction Of Arrivals (DOA’s) estimation techniques in modem radio
reconnaissance systems are based on eigen decomposition off-line analysis of received signals
covariance matrix. The common requirement to these techniques is that the array of sensors should
be linear and uniform. In many applications, the geometry of the array may be subject to
deformation in the field, or the elements are randomly distributed. In these cases, an approach for
processing the signal of arbitrary array geometry is necessary. This approach must be general to
include the cases of elliptical arrays on the hull of ships, V-and Y shaped arrays of radio
reconnaissance systems, and others. For alleviating the problem, which arises with the arbitrary
arrays, a pre processing technique to the received signal has been recently applied in a form
convenient only for planar structure.
This dissertation introduces the pre-processing technique to the signal of arbitrary array in a
form that can be applied to non-planar structure. The developed computer code applies
Multiple Signal
Classification (MUSIC) estimator m conjunction with the spatial smoothing technique. The
simulation results discuss the performance of the MUSIC estimator in the two cases of uniform and
arbitrary arrays. It also discusses the application of the pre-processing technique to V-& Y
Shaped arrays.
Direction finding of radio activities using an arbitrary array conformed to a spherical surface is
described. It is based on a transformation of the signal of the arbitrary array into a uniform
linear virtual one. In this form,
a selected direction finding estimator can be applied in the environment
’ . . ’
of Incoherent/Coherent sources. The sources are considered to be essentially narrow band.