On the Superresolution of Microwave Scanning Radiometer Measurements

Abstract

Microwave radiometer measurements are exploited to extract important geophysical information. Although it is beneficial to merge different frequency channels, it requires extra effort to refer all measurements to a common spatial resolution. Therefore, the capability to enhance the spatial resolution of a single channel is of special interest. In this study, a new numerical procedure to get superresolution microwave scanning radiometer measurements is presented. The approach is physically based on the occurrence of multiple partially correlated measurements. Mathematically, the approach is equivalent to a linear inversion problem, and its solution is pursued by means of a superresolution numerical procedure based on the Tikhonov regularization method. A set of numerical examples illustrates the results of the study in which hypothetical scanning microwave radiometer sensor configuration and reference test cases have been considered.