Adaptive algorithm based on FFT for frequency estimation

The performance of DFT (Discrete Fourier Transform) algorithm about frequency estimation is obviously declined, because the DFT spectrums have the problem of energy leakage and picket fence effect. Aiming at this problem, an adaptive algorithm based on FFT for frequency estimation is presented. In this paper we analysis the performance of Rife algorithm and point out that the perfor mance is poor when the true frequency is much close to quantized frequency of DFT. The performance of segmented-FFT (Fast Fourier Transform) phase difference method is also analyzed, and we point out that the performance is poor when the phase difference of the maximum spectral line is big. This adaptive algorithm is a combination of these two algorithms. The adaptive algorithm not only retains the advantages of the two algorithms, but also effectively makes up for the shortcomings of the two algorithms. The simulation results show that the anti-noise performance and stability of algorithm have significantly improved.