Please use this identifier to cite or link to this item:
|Title:||Development of a Near-Field Microwave Imaging System|
|Department:||Electrical and Computer Engineering|
|Abstract:||This thesis presents the results of an ongoing development of a near-field microwave imaging system. The thesis mainly focuses on the image reconstruction algorithms and data processing. Two linear-inversion methods, block circulant with circulant blocks (BCCB) scattered-power mapping (SPM) and convolution-based SPM, have been proposed. Both methods are general and efficient in solving the linear inverse problem. The images are reconstructed in quasi-real time with the BCCB SPM and in real time with the convolution-based SPM. A new method of building the SPM system matrix formed by the calibration object power maps is proposed. It allows for a reduced number of calibration measurements. BCCB SPM and convolution-based SPM are intended as a tool to solve weak-scattering problems or as a linear-inversion module within nonlinear iterative reconstruction. An algorithm has been developed for the de-noising of S-parameter data used in microwave imaging. It enables the efficient estimation of the noise-free signal component and its separation from the noise component in 2D-scan data sets. The proposed algorithm offers several benefits in imaging. First, it can suppress noise and uncertainties in the data used as input to the reconstruction algorithms. Such noise and uncertainties lead to image artifacts and errors. Second, the condition number of the BCCB system matrix improves as a result of the de-noising preprocessing of the raw data. The algorithm can also be used to quantify the imaging system’s dynamic range. Finally, it allows estimating the signal-to-noise ratio of a particular data set. Another development is concerned with a novel calibration strategy for near-field quantitative linear-inversion methods. It employs a metallic scattering probe embedded in the background. The biggest advantage of the calibration with metallic scattering probe is the target-independent quantitative accuracy. Also, full polarimetric information about the incident field can be readily obtained. An axial-null illumination has also been proposed to simplify the calibration of microwave imaging systems. Such illumination also enhances the spatial resolution. It can be achieved with various array configurations but a minimum of two transmitting antennas are required. Due to the intrinsic antisymmetry of the axial-null illumination, the baseline signals are suppressed down to the noise level of the measurement system. Therefore, the most important advantage of the proposed imaging setup is that it eliminates the need for background (or baseline) measurements. The discussed improvements are expected to occur for any imaging technique that exploits wave-like physical fields.|
|Appears in Collections:||Open Access Dissertations and Theses|
Files in This Item:
|Shumakov_Denys_S_201708_PhD.pdf||5.72 MB||Adobe PDF||View/Open|
Items in MacSphere are protected by copyright, with all rights reserved, unless otherwise indicated.