A risk minimization framework for channel estimation in OFDM systems

Abstract

We address the problem of channel estimation for cyclic-prefix (CP) Orthogonal Frequency Division Multiplexing (OFDM) systems. We focus on situations wherein knowledge of channel statistics are not available a priori, and model the channel as a vector of unknown deterministic parameters. Since computing the mean-square error (MSE) is not practicable in such a scenario, we propose a novel technique using Stein's lemma to obtain an unbiased estimate of the mean-square error, namely Stein's unbiased risk estimate (SURE). The channel estimate is obtained from noisy observations using linear and nonlinear denoising functions, whose parameters are chosen to minimize SURE. Based on computer simulations, it is shown that using SURE-based channel estimates for equalization offers an improvement in signal-to-noise performance over existing channel estimation schemes.