High-Resolution Tomographic Radar System for the Characterization of Low-Permittivity Materials

Nondestructive testing methods are an integral part of the quality control of many manufactured goods. This thesis introduces a novel high-resolution tomographic radar system intended for the characterization of low-permittivity materials, such as gaseous substances or light-weight foam.
Radar sensors positioned around an area-of-interest transmit electro-magnetic signals to probe the specimen under different incident angles. The 2-D permittivity distribution can then be reconstructed from the acquired propagation times of the radar signals.  
The system model is described in detail, and reliable signal processing methods are presented. Two complementing tomographic reconstruction algorithms - Tikhonov and total variation - are adapted to solve the inverse problem. A high-level system simulator is implemented, and the system's measurement uncertainty is analyzed.
The concept is validated using two different prototypes based on low-cost fully-integrated FMCW radar sensors operating in the mmW frequency range. Different binary gas mixtures as well as various low-density foam specimens are imaged, demonstrating the feasibility of the proposed system for a wide range of industrial applications.