Reconstruction algorithms for geophysical applications in noisy environments

Abstract
Remote determination of underground geophysical structures is a matter of considerable interest. Detecting such structures by scanning the area of interest with electromagnetic radiation is shown to reduce to the problem of solving an inconsistent system of equations for a spatial attenuation distribution. This paper examines the employment of reconstruction algorithms to solve such systems of equations, and it presents significant modifications to a standard algorithm that vastly improves its performance in the presence of noisy data. The process of in situ coal gasification is taken as an example to illustrate the superior performance of the modified algorithm.