An energy-efficient power allocation scheme for massive MIMO systems with imperfect CSI

Abstract As a design goal of green wireless communication, it is an important consideration to maximize energy efficiency of a Massive multiple-input multiple-output (MIMO) downlink system. By considering the case where the base station and all the users have the imperfect channel state information, a power allocation optimization problem is formulated to achieve the maximization of energy efficiency. Also, we derive the tractable closed-form expressions of lower bound on the achievable downlink rate utilizing maximum-ratio transmission (MRT) and zero-forcing precoding. Since the proposed optimization problem is non-convex, we strive to find a sub-optimal power allocation scheme. By decomposing the expression of the non-convex objective function, an energy-efficient power allocation algorithm that maximizes the energy efficiency is developed based on the difference of convex programming. Numerical results verify the tightness of the derived closed-form expressions, and show the performance improvement of the proposed energy-efficient power allocation scheme over the existing power allocation schemes.