Application of Swingler's method for analysis of multicomponent exponentials with special attention to non-equispaced data

Abstract

Swingler enhanced the work of Gardner to provide an elegant deconvolution method by which multiple summed exponential components might be resolved within time-domain data. Nevertheless, the application of the method remains limited owing to subtle complications that discourage many users. We present a tutorial and extend the approach to handle non-equispaced data. Finally the method's limits are identified in the case of closely-spaced exponential components with added input noise.