A New Synthetic Frequency Estimation Method of Sinusoid Based on Interpolated FFT

Abstract

A new frequency estimator of sinusoid based on interpolated FFT amplitudes is proposed. The coarse frequency estimation is made by the position of the maximum DFT amplitude. Then the fine estimation is made by utilizing the amplitudes of two DFT coefficients. Simulation results show that the root mean square errors of frequency estimation of the proposed estimator and Fang's estimator are close to the Cramer-Rao Lower Bound when the signal frequency locates in different regions between two neighboring FFT spectral lines. To further improve the estimator's performance, a synthetic frequency estimator is constructed. Simulation results show that the root mean square error of synthetic estimator is close to the Cramer-Rao Lower Bound when the signal frequency locates anywhere between two neighboring FFT spectral lines. The synthetic method has low computation complexity and outperforms the existing estimators.