Iterative channel estimation for turbo equalization over MIMO double selective fading channels

A novel pilot symbol assisted modulation (PSAM) based iterative channel estimation method over time and frequency selective fading channels is proposed, applied in conjunction with multiple-input-multiple-output (MIMO) turbo equalization. Initial channel estimates are derived from optimal training by maximum likelihood (ML) acquisition and subsequent Wiener filtering. In the iterative channel estimator, the MIMO frequency selective channel is decoupled into multiple single-input-single-output (SISO) flat fading sub-channels by using the soft decisions from the decoder to cancel both the inter-symbol-interference (ISI) and the inter-channel-interference (ICI) from the receive signals. Then refined channel estimates of individual MIMO multipath taps can be obtained independently from the corresponding virtual SISO flat fading channel. Simulation results show bit error ratio (BER) performance gains and normalized mean squared error (MSE) improvements achieved by the iterative channel estimation scheme over those of the non-iterative scheme for normalized fade rates f_dT_s=0.01 and f_dT_s=0.005. The effect of the suboptimal decoupling by the interference cancellation in the iterative estimator is investigated by comparing to the optimal decoupling situation.