Fundamental Frequency Estimation using the Shift-Invariance Property

In this paper, we propose a method for the estimation of the fundamental frequency of a periodic waveform based on the shift-invariance property of the sinusoidal signal model known from ESPRIT. An advantage of the proposed method is that the model order and the fundamental frequency can be found jointly and that the method does not depend on the density of the observation noise but rather on it being approximately white. Also, the cost function is observed to very smooth as compared to that of MUSIC. The method is shown in simulations to have good performance with the root mean square estimation error approaching the Cramer-Rao lower bound.