Bayesian Multiple Estimation in Flat Rician Fading MIMO Channels

In this paper, the performance of the singleestimation (SE) and multiple-estimation (ME) is investigated in multiple-input multiple-output (MIMO) Rician flat fading channels using the traditional least squares (LS) estimator and the Bayesian minimum mean square error (MMSE) estimator. The closed form equations are obtained for mean square error (MSE) of the estimators in SE and ME cases under optimal training. In ME case, the optimal weight coefficients are achieved for both estimators. Analytical and numerical results show that the LS estimator has lower error in the case of ME than SE. Moreover, it is seen that the performance of MMSE channel estimator in the ME case is better than SE particularly at high signal to noise ratios (SNRs). Furthermore, it is shown that this estimator is more appropriate for the channels with weak line of sight (LOS) propagation paths and/or the low correlations.