Analysis of multiexponential functions without a hypothesis as to the number of components

The analysis of a curve as a sum of exponentials is one of the most common problems encountered in experimental 'curve-fitting' in such diverse areas as electrophysiology, chemical kinetics, electrical engineering and nuclear science. In this paper, a new method for the analysis of multicomponent exponential functions (either real and/or complex) is presented. This method is based upon the combined use of the Laplace transform and of Padé approximants. As compared with approximation procedures, it does not require an a priori hypothesis as to the number n of components, which is an output of the analysis1. It thus becomes possible, under realistic numerical conditions, to address the problem of unambiguous detection of the exponential components. Performance comparisons show that several practical limitations reported in previous works are overcome. Detailed derivations relevant to the fundamental basis of the method described here are presented in ref. 2 and further developments pertaining to numerical analysis aspects may be found in ref. 13.