Iterative CZT-based frequency offset estimation for frequency-selective channels

In this paper, we present an accurate frequency offset estimation method for frequency-selective channels. Through the iterative use of the chirp z-transform (CZT) algorithm, an accurate frequency offset estimator is proposed, approaching the Cramer-Rao bound (CRB) even at low signal-to-noise ratio (SNR). Further, the estimation can he achieved within one block of training sequence, thus avoiding the transmission of repetitive known blocks as is usually required in many conventional methods. Meanwhile, the overall complexity is acceptable. More importantly, the CZT operation can utilize the fast Fourier transform (FFT) structure that is favourable for digital signal processor (DSP) implementation. Simulation results show that two or at most three iterations of the CZT computation are sufficient for an accurate frequency offset estimation in the SNR range from 0 dB to 30 dB.