| Abstract | This talk is aimed at newer practitioners of the EXAFS technique to provide a better understanding of the technique's capabilities and limitations, and to provide a foundation for several of the subsequent talks in this session. I will give a pedagogical overview of some fundamental aspects of EXAFS data analysis that are often not understood or are under appreciated. Basic concepts in Fourier analysis will be reviewed and applied to the EXAFS problem, particularly for understanding real information content, degrees of freedom, resolution, etc. Although the EXAFS technique is ultimately limited by systematic error, I will also discuss the effects of random errors. By working only with simulated spectra with Gaussian noise, I will demonstrate the applicability of several fundamental Fourier concepts, especially those relating to the degrees of freedom. The efficacy of standard error analysis will be demonstrated. Finally, these ideas will be applied to real data, where the role of systematic error will be quantified. |
