Time reversal based signal processing techniques for ultrawideband electromagnetic sensing in random media

This dissertation focuses on the application and development of time reversal (TR) based signal processing techniques for ultrawideband (UWB) electromagnetic waves in random media. TR techniques exploit the time reversal invariance of the wave equation to take advantage of the retransmitted (time-reversed) fields for improved imaging and focusing. This is because retransmitted signals propagate backwards through the same medium and undergo similar reflection, refraction and multiple scattering that they underwent during the forward propagation, resulting in focusing around the initial source locations. If multiple scattering occurs in the intervening media, refocusing resolution can overcome the classical diffraction limit (i.e. when no multiple scattering is present) that characterizes superresolution, a somewhat counterintuitive result. With the success of initial TR experiments in acoustics, there has been a strong interest in the application of TR methods using radio frequency electromagnetic (EM) waves. It is also the motivation of this dissertation to develop TR techniques for UWB electromagnetic waves. We start by investigating the superresolution effects of time-reversed UWB EM waves under continuous random background media and examine their dependency on the first- and second-order statistics. In contrast to the acoustic case, polarimetric TR exploiting the polarization of the EM waves is also analyzed to observe the depolarization effects on the refocusing of time-reversed EM waves. We continue with the application of TR in lossy and dispersive media (such as soil or biological tissues) where the TR invariance is broken and yields a performance degradation. We introduce a physical compensation technique based upon both space and frequency dependent inverse filters to improve the performance of TR techniques for homogeneous and random dispersive media. We also develop a physical full time-domain selective focusing method on the desired scatterers in the presence of others. The method is applied both in homogeneous and random media. Both narrowband and UWB TR-based imaging techniques for detection and localization of distinct scatterers in inhomogeneous background media are investigated under several perturbations. Specifically, we investigate the effects due to clutter, noise, dispersion and losses. Moving TRAs and restrictions on array elements on these TR-based imaging methods are also considered. Finally, we introduce a new TR-based imaging functional based on the simultaneous utilization of spatial and UWB frequency data to obtain a novel method for UWB imaging of embedded scatterers in homogeneous and random media.;As can be understood, although we consider typical subsurface sensing scenarios in presence of inhomogeneous and dispersive soil models, the same algorithms can be adapted to different applications, as is the case with microwave breast cancer detection or nondestructive testing.