Optimization Study of Frequency Estimation Combining Maximum Likelihood Estimation and Fast Fourier Transform

In noisy environments, frequency estimation faces greater challenges, and traditional methods have poor accuracy in the presence of high noise or signal irregularities. Therefore, improving the accuracy of frequency estimation in noisy environments is a key issue in current research. This study combines the maximum likelihood estimation (MLE) and fast Fourier transform (FFT) methods, utilizing the periodic characteristics of the signal for frequency estimation without relying on the signal's amplitude and phase information. With a true frequency of 30 MHz, experiments using different flight period data show that the MLE-FFT method can obtain stable frequency estimation results of 41.000 MHz. This research provides important theoretical support and practical application references for frequency estimation in complex signal analysis processes, indicating that the joint application of FFT and MLE methods can effectively achieve high-precision estimation of signal frequency.