Joint Spatial and Frequency Proportional Fairness Scheduling for MIMO OFDMA Downlink

Abstract

For multiuser MIMO system, zero forcing beamforming (ZFB) can be adopted to exploit the multiuser spatial multiplexing gain and diversity gain in downlink by spatial multiuser scheduling. On the other hand, multiuser joint subcarrier allocation can obtain multiuser diversity gain in frequency domain. In this paper, a novel joint spatial and frequency proportional fairness (SFPF) scheduler is proposed to achieve the spatial-frequency multiuser diversity gain for multiuser MIMO OFDMA based LTE TDD system while guarantees the multiuser fairness. To approach the multiuser capacity, multiuser ZFB (Mu-ZFB) is adopted for spatial multiplexing of multiuser on the same subcarrier. To simplify the antenna selection of Mu-ZFB, three simplified multiuser antenna selection algorithms are proposed, optimal, random and suboptimal antenna selection and their performance is evaluated in a single cell scenario. The optimal algorithm can achieve the best performance but has the highest complexity. The random algorithm has the worst performance but the lowest complexity. The performance comparison in a multi-cell scenario shows that three algorithms of SFPF can obtain a similar system throughput when guarantee the user fairness. While the suboptimal algorithm is a tradeoff between the performance and the complexity.