Multidimensional peak searching algorithm for low-statistics nuclear spectra

In the paper algorithms for peak searching in one-, two-, three-, and multidimensional spectra (histograms) are derived. They are based on Markov chain smoothing. Though they can be used for any type of nuclear spectra their importance is significant mainly for low-statistics spectra, where standard algorithms fail. The derived procedures are simple and therefore have a low cost in terms of computing time.