On the interpretation of diffusion battery data

Abstract

This paper presents a new procedure for reconstructing a submicron aerosol particle size distribution function (PSDF) from a finite number of noisy diffusion battery measurements. This illposed problem is solved by utilizing a maximum entropy approach to select, as the solution, that unique PSDF which maximizes the Shannon-Jaynes entropy subject to the constraints imposed by the given measurements. The utility of the method is demonstrated by a number of numerical examples with both noise-free and noisy synthetic diffusion battery data. This new method is compared with two existing inversion procedures and is shown to have superior performance in almost all of the simulated data examples tested.