Also, it seems like the author might not be aware that the FT of a Gaussian is another Gaussian. I implemented this method (Timmer and Konig) a few months ago,

Also, it seems like the author might not be aware that the FT of a Gaussian is another Gaussian. I implemented this method (Timmer and Konig) a few months ago, and I think there are some other statistical tricks that work out nice for Gaussians.