A Two-Stage Pilot-Assisted CFO Estimation Scheme for OFDM Signals

Abstract — In this paper, a pilot-assisted two-stage carrier frequency offset (CFO) estimation algorithm is proposed for OFDM signals. The proposed scheme performs a coarse search over a window of possible carrier offsets to obtain an initial estimate of the CFO. Then, a fine search is carried over a much smaller search window to obtain better accuracy of the carrier frequency offset. The pilot signal used in this algorithm results in a slowly varying correlation function with peaks at the correct offset. This is exploited by the proposed algorithm to reduce the search window for the second stage. Simulation results presented in this paper show that a significant improvement in the mean-square error performance is achieved by using this two-stage approach without increasing the complexity of implementation compared to single-stage conventional algorithms. It is also demonstrated that the proposed scheme outperforms the conventional cyclic-prefix based CFO estimation algorithm under both flat and frequency-selective Rayleigh fading channel conditions