Energy-Efficient Scheduling for Downlink Multi-User MIMO

Multi-user MIMO is the enabling technology forLTE-Advanced systems to meet IMT-Advanced targets. The gainof multi-user MIMO is achieved partially through advanced usergrouping,user-scheduling, and precoding. Traditionally, multiuserMIMO scheduling focuses solely on spectral-efficiency [1].That is, the scheduler will strike to balance the cell-edge userspectral-efficiency as well as the cell-average spectral-efficiency.Similar to spectral-efficiency, energy-efficiency is becoming increasinglyimportant for wireless communications. The energyefficiency is measured by a classical measure, “throughput perJoule”, while both RF transmit power and device electronic circuitpower consumptions are considered. In this paper, an energyefficientproportional-fair scheduling is proposed for downlinkmulti-user MIMO systems. To specific, the scheduling algorithmis proposed to balance cell-edge energy-efficiency and the cellaverageenergy-efficiency. The energy-efficient proportional-fairmetric is defined and the optimal power allocation maximizingthe performance measure is identified. System level evaluationsuggests that multi-user MIMO could improve the energyefficiencyof a wireless communication system significantly.QC 20130215low-energy wireless system