Performance comparison of the semi-blind channel estimation methods of MIMO communication system

In this paper, we present semi-blind channel estimation (SBCE) methods using two deterministic blind algorithms: Subspace (SS) approach and Least-squares smoothing (LSS) approach. A cost function is design by combining conventional training based method with quadratic blind criterion. As optimization criterion, the channel normalized mean square error (NMSE) is chosen and compares the performance of semi-blind method using subspace and Least squares smoothing criterion with the conventional training based method as a function of weighting factor (α).