Date Added: Dec 2011
A novel iterative channel estimation algorithm for Orthogonal Frequency-Division Multiplexing (OFDM) systems in fast time-varying environment is proposed. Instead of directly estimate the original channel taps, the proposed estimation focus on the rest channel part from the original channel taps subtracted by the prior estimation results in the absence of any channel information. The initial results are obtained by time-domain Least Squares (LS) estimation by inserting zeros-padded training symbols between adjacent OFDM symbols. Monte Carlo simulations demonstrate the superiority of the proposed algorithm after a few iterations over the conventional methods based on Complex-Exponential Basis Expansion Model (CE-BEM) or Generalized Complex-Exponential Basis Expansion Model (GCEBEM).