摘要

In virtual multiple-input-multiple-output (V-MIMO) systems, employing a user-pairing algorithm can significantly improve system capacity and spectral efficiency using multiuser diversity. The optimal pairing algorithm can achieve optimal system performance, but it is not feasible to implement in practice due to its huge complexity. To solve the problem, an efficient suboptimal pairing algorithm with lower complexity is proposed, which maximizes the system throughput by selecting users one by one. Furthermore, to achieve a good balance between throughput and user fairness, a novel pairing algorithm is derived, which combines the advantages of the suboptimal pairing algorithm and proportional fair (PF) criteria. Simulation results show that the proposed algorithm achieves better fairness and higher system throughput than the double PF (D-PF) algorithm, and has much lower complexity than the optimal pairing algorithm.