Accurate sine-wave frequency estimation by means of an interpolated DTFT algorithm
Résumé
An Interpolated Discrete Time Fourier Transform (IpDTFT) algorithm for sine-wave frequency estimation is proposed in this paper. It generalizes the classical interpolated Discrete Fourier Transform (IpDFT) algorithm by interpolating two DTFT spectrum samples located one frequency bin apart. The influence on the estimated frequency of the spectral image component is investigated in the case when the acquired sine-wave samples are weighted by a Maximum Sidelobe Decay (MSD) window. An analytical expression for the estimation error due to the spectral image component is derived. Leveraging on that expression an iterative procedure for the reduction of the effect of spectral image component on the estimated frequency is proposed. The accuracies of the proposed procedure and other state-of-the-art interpolated DFT algorithms are compared by means of both computer simulations and experimental results. It is shown that the proposed procedure can be advantageously adopted when the number of acquired sine-wave cycles is small.