Skip to Main Content
We consider a block fading frequency selective multi-input multi-output (MIMO) channel in additive white Gaussian noise (AWGN). The channel input is a training vector superimposed on a linearly precoded vector of Gaussian symbols. To achieve a better performance over the conventional least-squares (LS), we utilize the linear mean square error (LMMSE) symbol estimate to improve the initial LS estimate and update the symbol estimation accordingly. We provide the guidelines to design training which minimizes the MSE of the initial LS estimate.