On Doubly Dispersive Channel Estimation for Pilot-Aided Pulse-Shaped Multi-Carrier Modulation

In this paper, we propose several methods for the pilot-aided estimation of significant ICI coefficients resulting from pulse-shaped multicarrier modulation (PS-MCM) over DD channels. Specifically, we outline Wiener and reduced-rank (RR) Wiener estimation schemes that leverage statistical channel structure, as well as deterministic least-squares (LS) schemes based on basis expansion modeling (BEM). We then report the results of a numerical study which suggests that RR Wiener estimation outperforms LS estimation based on polynomial and oversampled complex exponential BEM, even under significant statistical mismatch. In addition, the RR Wiener estimator is computationally cheaper than the LS-BEM techniques. These findings have implications on the practical design of PS-MCM channel estimation schemes.