Minimum interference blind time-offset estimation for OFDM systems

This paper proposes a novel blind time-offset (TO) estimation scheme for OFDM systems. Based on the observation that a time-offset almost surely gives rise to interblock interference (IBI), our TO estimation is accomplished by minimizing the IBI. Borrowing the idea of differential decoding, the need for channel state information is circumvented via joint processing of multiple received blocks. The proposed TO estimator is shown capable of offering reliable TO estimation with a small amount of received data. Unlike some existing schemes, it performs well in both frequency-flat and frequency-selective fading channels. More importantly, the proposed TO estimator is consistent, and guarantees identifiability of the TO with probability one. In addition, since most required computation can be done by using the FFT, it enjoys low computational complexity. The merits of our blind TO estimator are investigated by both theoretical analysis and corroborating simulations.