Unified performance analysis of blind feedforward timing estimation

By exploiting an unified cyclostationary (CS) statistics based framework, this paper develops a rigorous and thorough performance analysis of blind feedforward timing estimators for linear modulations. Within the introduced CS-framework, it is shown that several estimators proposed in the literature can be asymptotically interpreted as Maximum Likelihood (ML) estimators, and the timing estimator proposed by Oerder and Meyr in [6] achieves the lowest asymptotic (large sample) variance in the class of estimators which exploit all the second-order statistics of the received signal, and its performance is insensitive to the oversampling rate P as soon as P ≥ 3. Further, based on the presented theoretical analysis, we derive an Asymptotically Best Consistent (ABC) estimator, which exhibits asymptotically the best performance among all the existing blind feedforward timing estimators, especially when dealing with strongly bandlimited signaling.