Ionospheric imaging using computerized tomography

Computerized tomography (CT) techniques can be used to produce a two-dimensional image of the electron density in the ionosphere. The necessary data are transionospheric satellite beacon total electron content (TEC) data recorded simultaneously at multiple ground stations.The ionospheric imaging case presents a difficult problem due to large amounts of missing data. This is a consequence of the locations of the transmitter (in orbit) and receivers (ground-based) and causes the reconstruction algorithm to fail to correctly reconstruct the background density profile. Despite this limitation, a method has been developed which successfully reconstructs the irregularities and variations in the profile.A computer program simulates the data collection and image reconstruction process. This allows the method to be tested with several electron density models, transmitter and receiver locations, and noniterative and iterative reconstruction algorithms. Simulations are performed assuming a 1000-km-altitude polar-orbiting satellite and several ground stations. The imaged region is 3500 km wide by 800 km high.A new algorithm, which is a modification of the discrete backprojection algorithm, is developed for use with nonuniform sampling geometries. It produces good quality results in cases where the unmodified algorithm fails to produce a usable result.Simulations are performed on both test images and realistic ionospheric models in order to demonstrate the capabilities and limitations of this technique. Through these simulations, a combination of modified CT algorithms was found to produce good results. Because of the incomplete data used for the reconstruction, the initial estimate used for iterative algorithms is very important. Methods of incorporating other data, such as height profiles obtained from an ionosonde, into the reconstruction procedure to eliminate this problem are suggested.Chapter 1 contains an introduction, the statement of the problem, motivation, and similar research. Chapter 2 contains TEC measurement theory, CT theory, and a description of several CT algorithms. Chapter 3 contains a description of the simulation procedure, the models used, and the simulation results. Chapter 4 contains a discussion of the results, comments on the program, suggestions for improvements, and conclusions. The Appendix documents the simulation program, presenting details of the data structures, and providing algorithm listings.U of I OnlyETDs are only available to UIUC Users without author permissio