Algorithm for sine wave cure fit based on frequency precise estimation

In the process of curve fitting, the unknown relationship between the data sampling rate and the frequency of the measured signal as well as signal's frequency fluctuations can cause remarkable error. In this paper, a sine curve fitting algorithm with frequency precise estimation is proposed to solve this problem. Firstly, ellipse fitting algorithm is used to estimate the measured signal frequency. On the basis of this, the three-parameter sine fitting calculation based on the least squares method is applied to the sampling data, and the accuracy of the sine curve fitting result is improved. The experimental results show that the algorithm proposed in this paper can avoid the problem of frequency fluctuation and error caused by the unknown relationship between the sampling frequencies. Moreover, the fitting has a good randomness after frequency estimation, and the fitting result has nothing to do with the signal amplitude. Compared without frequency accurate estimation, the fitting error is greatly reduced and the fitting precision is higher.