Noise reduction of PGNAA spectrum using empirical mode decomposition technique

Abstract

Prompt gamma neutron activation analysis (PGNAA) is frequently used in on–line analysis of raw materials in cement and coal industries which normally incorporates bismuth germanate (BGO) or sodium iodide (NaI(Tl)) scintillators. The quantitative results are basically undertaken through full–spectrum analysis method which requires an efficient noise reduction procedure to optimize PGNAA data. Here in this study, the empirical mode decomposition (EMD) technique has been used to reduce the noise level of PGNAA spectra of a neutron–irradiated NaCl·H2O solution and then the results have been compared with those of wavelet technique. The mean square errors (MSEs) comparisons confirm that both EMD and wavelet techniques can be efficiently used for noise reduction purposes, however, this study recommends the EMD technique for its independency of basic functions, simplicity and spectrum data-based operation.