Iterative B-spline channel estimation for fast flat fading channels

A novel B-spline iterative channel estimation technique over fast flat fading channels is proposed. Both local linear and parabolic splines are considered. The optimal sampling interval is found by simulations and then approximated by a simple equation. Comparisons with the Wiener filtering approach in mean square error (MSE), bit error rate (BER) and complexity are given. The BER performance of the iterative receiver with the proposed estimators is very close, within 0.3 dB for BER = 10^-4 of that of the Wiener estimator for fading rates up to f^dT^s = 0.02. However, the proposed estimators require only a few multiplications per symbol per iteration which is only a small fraction of that of the Wiener filter.