Files in this item
|(no description provided)|
|Title:||Tapping space and spacing taps: Topics in radar-astronomy imaging and sparse filter design|
|Author(s):||Webb, Jennifer Lois Harmon|
|Doctoral Committee Chair(s):||Munson, David C., Jr.|
|Department / Program:||Electrical and Computer Engineering|
|Degree Granting Institution:||University of Illinois at Urbana-Champaign|
|Subject(s):||Engineering, Electronics and Electrical|
|Abstract:||This dissertation addresses topics in the areas of synthetic aperture radar (SAR) and finite impulse-response (FIR) digital filter design.
This investigation of SAR was based on a tomographic viewpoint. An existing form of spotlight-mode SAR processing was applied to a radar-astronomy application, forming images of the moon's surface from radar-astronomy data collected at Arecibo Observatory. In doing so, high-quality images were obtained while significantly reducing the number of computations required. It was found that a slight degradation was caused by arcing in the radar path.
This discovery led to a detailed investigation of the impact of radar maneuvers on SAR image formation. A method for improving image quality was developed, based on a piecewise-linear model for the radar path. To determine the number of segments required, and the appropriateness of this model, constraints were derived for the length of the linear path segments. The method was demonstrated for 1-D images with point scatterers and extended targets.
Another study, motivated by the problem of imaging asteroids, involved the formation of 3-D SAR images using limited data. When the surface reflectivity is known to be concentrated on the surface, it is possible to cast this problem into the framework of an angle-of-arrival or spectral-estimation problem. Thus, techniques such as ESPRIT and RELAX can be applied to form the sparse 3-D image. The image-formation method was demonstrated through simulation.
Other work deals with FIR digital filter design. These studies involved designing filters for which some of the tap weights are constrained to be zero. First, cases were considered where the locations of the zeroed tap weights are known; linear programming was used to optimize the remaining tap weights. If the goal is to reduce the number of multiplications, the locations of the zeroed tap weights must be determined first. For a class of filters, these locations were determined and a sparse direct-form filter was designed using linear programming. A second new design method was developed for filters implemented with cascades of sparse and nonsparse sections.
|Rights Information:||Copyright 1995 Webb, Jennifer Lois Harmon|
|Date Available in IDEALS:||2011-05-07|
|Identifier in Online Catalog:||AAI9624533|
This item appears in the following Collection(s)
Graduate Dissertations and Theses at Illinois
Graduate Theses and Dissertations at Illinois
Dissertations and Theses - Electrical and Computer Engineering
Dissertations and Theses in Electrical and Computer Engineering