THREE-DIMENSIONAL IMAGING OF COHERENT AND INCOHERENT SOURCES UTILIZING WAVEVECTOR DIVERSITY.

摘要

This dissertation unifies the class of imaging techniques which use spectral degrees of freedom to obtain high resolution three-dimensional images of spatially coherent and incoherent broadband sources. We show that the basis concepts of broadband coherent three-dimensional imaging by wavevector diversity N. Farhat et. al., J. Radio Science, 19, 1347, 1984 are extendable to projection and tomographic imaging of three-dimensional, spatially incoherent, noise emitting bodies. The extension is based on the duality of the propagation and scattering behavior of the complex field amplitude, and the mutual coherence function.;Two equivalent methods are described for accessing the three-dimensional Fourier space of spatially incoherent broadband sources. The first system is based on direct measurement of the mutual coherence function, using charge coupled analog delay elements. The second system estimates the cross spectral density of the field from spectrally selective cross-correlation measurements using dual, switched capacitor filters under computer control.;The results of a series of imaging experiments, the first of their kind, will be described for a discrete source consisting of three acoustic speakers excited by amplified thermal noise. Tomographic images (slices) through object, obtained by application of the weighted projection-slice theorem to the three-dimensional Fourier data, are presented. The results represent the first demonstration of 3-D tomographic imaging of incoherent objects employing spectral degrees of freedom. Verification of the aperture synthesis by wavevector diversity for spatially incoherent sources leads to practical applications in 3-D radiometry, thermography, sonar, and noise analysis of machinery and has implications in remote sensing, surveillance, and emitter identification.;Microwave diversity imaging was the first application to employ spectral and angular degrees of freedom to obtain high resolution tomographic images. Three-dimensional images are presented in the form of tomographic parallel slices of the object, obtained from actual wavevector diversity data. Spectral, angular and polarization degrees of freedom have been utilized in addition to a priori knowledge of the object symmetry.