STLN-based channel estimation using superimposed training and first-order statistics

In this paper, Channel estimation using superimposed training and first-order statistics is considered. Information-induced interference matrix in channel estimation is of Toeplitz structure, which can be utilized for deconvolution of the system equation. A structured total least norm (STLN) approach is introduced to improve the estimation performance. Simulation results show the enhancement performance of the STLN estimator when compared with the LS, total least squares (TLS) and data least squares (DLS) estimators.