Energy efficient massive MIMO system design for smart grid communications

Communication technologies are critical in achieving potential advantages of smart gird (SG), as they enable electric utilities to interact with their devices and customers. This paper focuses on the integration of a massive multiple-input multiple-output (MIMO) technique into a SG communication architecture. Massive MIMO has the benefits of offering higher data rates, whereas operating a large number of antennas in practice could increase the system complexity and energy consumption. We propose to use antenna selection to preserve the gain provided by the large number of antennas, and investigate an energy efficient massive MIMO system design for SG communications. Specifically, we derive a closed-form asymptotic approximation to the system energy efficiency function in consideration of channel spatial correlation, which exhibits an excellent level of accuracy for a wide range of system dimensions in SG communication scenarios. Based on the accurate approximation, we propose a novel antenna selection scheme aiming at maximizing the system energy efficiency, using only the long-term channel statistics. Simulation results show that the proposed antenna selection scheme can always achieve an energy efficiency gain compared to other selection schemes or baseline systems without antenna selection, and thus is particularly valuable for enabling an energy efficient communication system of the SG.