DFT Frequency Estimation Threshold Level with Windowed Data

Discrete Fourier transform (DFT) based frequency estimation is an important topic and an often applied method in measurement science and signal processing in general. This is due to its optimality in terms of frequency estimation error, achieving the best possible performance given by the Cramer Rao lower bound (CRLB) asymptotically and the possibility to implement it computationally efficient via a fast Fourier transform (FFT). Below a certain level of signal-to-noise ratio (SNR), the so called threshold effect occurs, and the performance of the estimator decreases quickly. In this paper, we investigate the effect of windowing the data prior to the frequency estimation process on the threshold