A New MMSE Estimator Based on Signal Correlation Property

An improved minimum mean square error (MMSE) estimate method based on the diversity of correlation characteristic between the additive white noise and the signal is proposed in this presentation. The new estimator is constructed as a linear combination of the covariance matrices of the observed signal and its delay. The proofing and simulation indicate that the new estimator can achieve a better estimate than the traditional MMSE estimator.