A Maximum Likelihood Doppler Frequency Estimator for OFDM Systems

This paper derives a maximum likelihood Doppler frequency estimator for orthogonal frequency division multiplexing (OFDM) systems in time-varying multipath channels. The proposed scheme is a frequency-domain approach that utilizes pilot subcarriers, which are commonly implemented in most practical systems. Time-varying fading causes intercarrier interference (ICI) in OFDM systems. Thus, in the proposed estimator, the effect of ICI is taken into consideration with a proper model for accurate results. The estimator can be implemented using a finite impulse response (FIR) filter bank whose coefficients can be pre-calculated and stored in order to lower the computational complexity. We evaluate various methods to improve the estimation accuracy and analyze their complexity-performance tradeoffs. We also derive the Cramér-Rao bound and provide simulation results to quantify the performance of the proposed algorithm.