Title :
A comparison of the computational efficiency of three inverse scattering methods
Author :
Hesford, A.J. ; Chew, W.C.
Author_Institution :
Dept. of Electr. & Comput. Eng., Illinois Univ., Urbana, IL, USA
Abstract :
This paper presented three inverse scattering methods. Based on the provided data, it is clear that the distorted Born iterative method provides the most accurate reconstructions. However, the cost of such accuracy is computation time. Because this is a direct implementation of the method, one expects reduced times from a more clever design. While both the linear sampling method and diffraction tomography introduced significant errors into their reconstructions, diffraction tomography seems to be a better inversion method. Superior scalability and information about the electrical properties of scatterers makes tomographic reconstructions more useful than linear sampling method images. These results are consistent with the full analysis in Hesford (2004), as well as with those presented in Brandfass et al. (2001). In general, the linear sampling method is not capable of providing better images than diffraction tomography.
Keywords :
electromagnetic wave diffraction; electromagnetic wave scattering; inverse problems; iterative methods; sampling methods; tomography; diffraction tomography; distorted Born iterative method; inverse scattering methods; linear sampling method; reconstructions; Computational efficiency; Costs; Diffraction; Image reconstruction; Inverse problems; Iterative methods; Sampling methods; Scalability; Scattering; Tomography;
Conference_Titel :
Antennas and Propagation Society International Symposium, 2005 IEEE
Print_ISBN :
0-7803-8883-6
DOI :
10.1109/APS.2005.1551753