For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Scheduling Algorithms for Maximizing Throughput with Zero-Forcing Beamforming in a MIMO Wireless System
Augusto FORONDA Chikara OHTA Hisashi TAMAKI
IEICE TRANSACTIONS on Communications
Publication Date: 2008/06/01
Online ISSN: 1745-1345
Print ISSN: 0916-8516
Type of Manuscript: PAPER
Category: Wireless Communication Technologies
wireless network, multiple-input/multiple-output (MIMO), zero-forcing beamforming (ZFBF), multiuser diversity,
Full Text: PDF(618.7KB)>>
Dirty paper coding (DPC) is a strategy to achieve the region capacity of multiple input multiple output (MIMO) downlink channels and a DPC scheduler is throughput optimal if users are selected according to their queue states and current rates. However, DPC is difficult to implement in practical systems. One solution, zero-forcing beamforming (ZFBF) strategy has been proposed to achieve the same asymptotic sum rate capacity as that of DPC with an exhaustive search over the entire user set. Some suboptimal user group selection schedulers with reduced complexity based on ZFBF strategy (ZFBF-SUS) and proportional fair (PF) scheduling algorithm (PF-ZFBF) have also been proposed to enhance the throughput and fairness among the users, respectively. However, they are not throughput optimal, fairness and throughput decrease if each user queue length is different due to different users channel quality. Therefore, we propose two different scheduling algorithms: a throughput optimal scheduling algorithm (ZFBF-TO) and a reduced complexity scheduling algorithm (ZFBF-RC). Both are based on ZFBF strategy and, at every time slot, the scheduling algorithms have to select some users based on user channel quality, user queue length and orthogonality among users. Moreover, the proposed algorithms have to produce the rate allocation and power allocation for the selected users based on a modified water filling method. We analyze the schedulers complexity and numerical results show that ZFBF-RC provides throughput and fairness improvements compared to the ZFBF-SUS and PF-ZFBF scheduling algorithms.