[en] A novel auto-correlation function (ACF) method has been investigated for determining the oscillation frequency and the decay ratio in BWR stability analyses. The report describes not only the method but also documents comprehensively the used and developed FORTRAN codes. The neutron signals are band-pass filtered to separate the oscillation peak in the power spectral density (PSD) from background. Two linear second-order oscillation models are considered. The ACF of each model, corrected for signal filtering and with the inclusion of a background term under the peak in the PSD, is then least-squares fitted to the ACF estimated on the previously filtered neutron signals, in order to determine the oscillation frequency and the decay ratio. The procedures of filtering and ACF estimation use fast Fourier transform techniques with signal segmentation. Gliding 'short-time' ACF estimates along a signal record allow the evaluation of uncertainties. Some numerical results are given which have been obtained from neutron signal data offered by the recent Forsmark I and Forsmark II NEA benchmark project. They are compared with those from other benchmark participants using different other analysis methods. (author)