Deconvolution of fluorescence decays and estimation errors

摘要

The non-iterative Prony's method is discussed for the deconvolution of multi-exponential fluorescence decays. The performance of these algorithms in the case of a two-exponential decay process is evaluated, using a Monte-Carlo simulation, in terms of the estimation errors caused by the signal noise. The results which are presented show that the performance of Prony's method can be greatly improved with the selection of an optimized observation window width and a few algorithm-related parameters. Comparison between Prony's method and the Marquardt least-squared-error algorithm is also made, showing the performance of the former is close to that of the latter with a 98% reduction in calculation running time. The applications of Prony's algorithms in real-time, quasi-distributed temperature sensor systems are discussed and the experimental results are presented to justify the use of the algorithms in Prony's method in practical double exponential fluorescence decay analysis.