Abstract
Solving the inverse problem of obtaining field sources can be classified into following two methods. The first method consists of obtaining unique solution by imposing a certain constraint on the governing equation of the inverse problem. This is because an infinite number of possible solutions can satisfy the governing equation. In other words, the same local field can be generated by different field source distributions. The second method carries out examination of all the possible solutions, and their weight evaluation. The sampled pattern matching (SPM) method belongs to this method, while the minimum-norm method represents the former method. In this paper, comparison between the SPM and minimum-norm methods is carried out. As a result, it is clarified that the SPM method is superior to the minimum-norm method when the field source is distributed in the direction normal to a field measurement surface. Finally, a faster SPM method is proposed in order to realize a real-time analysis of the inverse problem. The validity of the faster SPM method is verified by experiments.
