Noise Influence on Estimation of Signal Parameter From the Phase Difference of Discrete Fourier Transforms

Estimation of multi-frequency signal parameters by frequency domain non-linear least squares

“…From different point of views, similar ideas are explored by Danthez and Giroussens [2], and Dai and Gretsch [27]. The statistical properties of the corresponding estimators show that this method has superior accuracy over the interpolationbased method [28].…”

“…Ratios between the non-linear least squares-based estimator variances and the interpolation- [22] and phase-difference-based ones [28] are shown in Figs. 2 and 3, as functions of a: It is observed that the former ones are all much smaller than the latter two ones.…”

“…Recently, Carbone et al [29] present a novel frequency domain least squares method, however, only a simplified linear model is employed to yield an improved amplitude estimator, while the frequency and phase estimators keep being the same as those reported in interpolation-based method due to the complexity of the comprehensive complex-valued model. Motivated by their work and our previous works [4,28], with two weak restrictions on the window function used, a concise frequency domain least squares-based parameter estimation model along with an efficient solution algorithm and detail performance analysis is presented in this paper. The remainder of this paper is organised as follows.…”

Anti-Noise Performance and Parameter Estimation Accuracy of FFT and FT Discrete Spectrum Correction

“…Qi Guo-qing [7] has investigated the frequency estimation accuracy of sinusoid on the interpolated FFT. Analysis of noise influence on the performance of phasedifference method is investigated by Li-Min Zhu [8] under condition that shifting points equal to FFT points. The antinoise performance and improvement measures of both window-length changing phase difference correction method and time-shifting correction method of phase difference in the presence of Gaussian white noise are presented by Yang Zhijian and Ding Kang [9].…”

Impact of harmonics on the interpolated DFT frequency estimator

“…The term δ is varied in the range (-0.5, 0.5) with a step of 0.04 cycles and the Hann 156 window is adopted. are determined by (13) and (14), respectively, using the harmonic parameters returned by the IpDFT …”