An Interpolation Algorithm for Discrete Fourier Transforms of Weighted Damped Sinusoidal Signals

Abstract

An algorithm for improving the accuracy of the estimation of parameters that characterize a multifrequency damped sinusoidal signal is presented. The core ideas of the algorithm is that the signal is weighted using Hann window before discrete Fourier transform and the frequencies, amplitudes, phases, and damping factors of the signal are obtained by frequency domain interpolation. It is shown that the purpose of improving the accuracy of parameter estimation is achieved using Hann window, which reduces long-range leakage and frequency-domain interpolation that eliminates the effect of the picket fence. The sensitivity analysis of the changing of the signal parameters, noise and spectral resolution, shows that both the effects of strong noise and spectral interference are considered, proving the reliability and high-accuracy parameters estimation in a number of engineering applications. Otherwise, the characteristic of efficiency computational and low memory demands are advantageously adopted for the poor computing resources situations. Extracting the impact signal of the rotor test bed using this method is illustrated. The result shows that the method can be an optional method for the fault features extraction of the mechanical impact.