Semi-Blind Channel Estimation based on subspace modeling for Multi-user Massive MIMO system

‎Channel estimation is an essential task to fully exploit the advantages of the massive MIMO systems‎. ‎In this paper‎, ‎we propose a semi-blind downlink channel estimation method for massive MIMO system‎. ‎We suggest a new modeling for the channel matrix subspace. Based on the low-rankness property, we have prposed an algorithm to estimate the channel matrix subspace. In the next step, using optimized pilots, we estimate the unknown parameters of the model and obtain more precise approximation of the channel matrix.

‎‎ ‎The simulation results confirm the superiority of the suggested method over its counterparts in various scenarios‎.