PILOT CONTAMINATION MITIGATION BASED ON INTERFERING USER'S ANGLE OF ARRIVAL IN MASSIVE MIMO SYSTEMS

A new approach to mitigate pilot contamination in massive multiple-input multiple-output (MIMO) systems is proposed in this paper. We consider two cells from the first tier of copilot cells of a cellular network where the base stations (BSs) are equipped with uniform linear arrays with hybrid beamforming adopted. We consider one cell as the cell of interest containing a typical desired user, and the other cell contains an interfering user sending data and contaminating pilot signals to the BS of the cell of interest. We derive a closed-form expression for the desired user’s achievable rate as a function of the interfering user’s angle of arrival (AoA). We model the ray propagation from the interfering user to the BS of the cell of interest and its related AoA as Gaussian distribution. Based on the model, we derive closed-form expressions for the pilot contamination level in the cell of interest and for the desired user’s data path gain estimation error due to pilot contamination. A perfect agreement is found between theoretical and Monte Carlo simulation results which show that when the interfering user’s AoA is increased the pilot contamination level is significantly minimized, the desired user’s data path gain estimation error also minimized, and hence its data rate is significantly increased. Moreover, we show in our analysis that the interfering user’s AoA can be effectively controlled and increased by reducing the copilot cells’ radius.