NEW PHYSICS-BASED INVERSE-SCATTERING TECHNIQUES FOR ULTRAWIDEBAND DISTRIBUTED SENSING

点击翻译按钮获取中文摘要

The objective of this research is to develop robust physics-based time-domain ultra-wideband inverse scattering techniques to detect obscured targets in random media using radio-frequency and microwave signals. The approach is based on an information-theoretic, Bayesian paradigm where the inverse scattering problem of detecting, imaging, and classifying obscured targets is treated as a conjunction of states of information that combine (1) a-priori information data with (2) (physics-based) forward problem solution and sensor data acquisition, to produce (3) the a posteriori state of information. The intellectual contributions of the proposed ultra-wideband inversion methodology involve (1) exploitation of super-resolution effects provided by spatial decorrelation of ultra-wideband signals in disordered media to yield imaging resolution beyond the classical diffraction limit (super-resolution) and (2) harnessing of time-reversal based synthetic back-propagation of ultra-wideband waves to provide frequency decorrelation in random media and, hence, statistically stable detection/imaging algorithms. The proposed physics-based inversion techniques will provide new capabilities for ultra-wideband sensors aimed at improving the way detection and imaging systems are employed at microwaves and radio-frequencies. The proposed research has the potential to benefit search-and-rescue operations where rescue teams must be able to locate hidden survivors under collapsed buildings, landslides, and other debris. It can also benefit surveillance systems for detecting objects and personnel behind walls (thru-wall imaging), under roofs, or around corners, and ground penetrating radar systems for humanitarian demining. The proposed research objectives are integrated with an education component aimed at providing a well-rounded training to students at The Ohio State University.