A semi-blind MIMO adaptive equalizer based on independent space constraint

This work presents a semi-blind independent space constrained flat channel equalization adaptive algorithm for a multiple input multiple output (MIMO) system. The proposed equalization method consists of two stages, namely the training stage, where a short training signal is employed, and a blind adaptation stage, where a blind adaptive algorithm is used to ensure that the demixing matrix is always a nonsingular matrix. In the first stage of equalization, orthogonal signals are transmitted in each transmit antenna. At the receiver, the training signal is used to evaluate channel´s state information (CSI), which is then projected to independent spaces. The blind stage of adaptation, which is based on the independent space constraint, has a relatively low computational complexity in comparison to other blind adaptive algorithms. Simulation results that have been provided verify that the algorithm is stable, after initialization using training signal.