Abstract
In this work, we propose a recursive semiblind channel estimation algorithm to mitigate the ill effects of pilot contamination in massive multi- user multiple-input multiple output (MU-MIMO) systems. In the proposed approach, we first project the received signals onto the subspace with minimal interference, where a low-complexity modified power method is employed to recursively determine the bases of this subspace. Specifically, a few pilot symbols are applied to generate initial channel estimate of the projected channel coefficients. Finally, a recursive algorithm is introduced to alternatively detect the data symbols and update the channel estimates. Notably, the proposed scheme requires neither cell cooperation nor pilot scheduling. Simulation results demonstrate the performance superiority compared with the existing works.