Noise influence on estimation of signal parameter from the phase difference of discrete Fourier transforms

This paper investigates the noise influence on the performance of the phase-difference-based signal parameter estimation method. The circular nature of the estimators that may be harmful to parameter estimation is explored, and a simple strategy to unwrap the circular estimators is proposed. Analytical expressions of the estimator variances, which allow both the evaluation of the result accuracy and the design of the measurement procedure. are derived, and they indicate that for noisy signals, comparing with the popular interpolation-based method, the phase-difference-based method provides superior accuracy in frequency and phase estimates, and approximately equal accuracy in amplitude estimates. Simulation results confirm the validity of the presented analysis. (C) 2002 Elsevier Science Ltd. All rights reserved.