Aspects of digital signal processing in radiometric remote sensing of geophysical variables

When geophysical variables are estimated from remotely sensed electromagnetic data, severe difficulties are often encountered because of the ill-posed nature of the problem, so that the use of effective processing becomes essential. The inversion of radiometric data in the retrieval of atmospheric temperature profiles is considered and the use of the Kalman filter algorithm is analyzer to assess its performance from the point of view of accuracy of estimates and spatial resolution. The sensitivity to experimental errors is pointed out and the use of adaptive filtering to reduce the inaccuracy of data is examined. With regard to spectra of radiometric fluctuations, the performance of recently proposed spectral estimation techniques, amenable to the originally proposed maximum entropy method, are considered. The estimate of line parameters to identify and follow the temporal evolution of atmospheric internal oscillations is discussed, as well as the accuracy of retrieval of the slopes of radiometric turbulent spectra.