A Fine Resolution Frequency Estimation Method for Noisy Signal and its Application

The frequency interpolation estimation method based on the discrete Fourier transform is susceptible to being affected by noise, resulting in interpolation error. The present work attempts to improve the frequency estimation accuracy while retaining the high efficiency, and consider its application in mechanical vibration engineering. A fine resolution frequency estimation method with the third-order frequency deviation estimation is proposed. The first deviation estimation is a coarse estimation, the results of the Rife and Quinn method are used as a coarse estimate of frequency deviation. Then the second estimation is deviation shifted, by judging whether the difference between the initially estimated frequency and the maximum spectral line frequency is located in the center of two adjacent quantization frequency points of DFT, the first estimated deviation is shifted to the effective spectral deviation range. Finally, the third estimation is accurately estimated, the maximum and minimum values of the effective spectrum range are calculated and interpolated. Simulation and experimental results show that the proposed method has high-frequency estimation accuracy. The proposed method is better than other frequency estimation methods under the same simulation and test conditions. The proposed method has excellent application in mechanical vibration fault diagnosis engineering.