Improved User Scheduling Strategies for Beam-domain Based 5G Systems

• Main Authors: HSIEH, CHI-AN, 謝騏安
• Other Authors: LIU, WEI-CHENG
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/50940714801556945107

碩士 === 國立中正大學 === 通訊工程研究所 === 104 === Using large-scale antenna in the next generation mobile communications systems is one of the most popular research topics in recent years. Using hundreds of antennas at the base station, a large number of users can be served on the same time/frequency resources simultaneously and therefore increasing system capacity. Beam-group Division Multiple Access (BgDMA) is such a system. There are three key features in BgDMA system. (i) Use the beam-channel processing technique in spatial domain to eliminate frequency selectivity. (ii) Users can take advantage of multiple-beam channel to achieve diversity gain and spatial multiplexing gain. (iii) Use low-complexity beam grouping to achieve the high-order MU-MIMO. In this thesis, we focus on the user scheduling for BgDMA MU-MIMO down-link systems in which user scheduling is done on user group basis. Compared with the existing methods, it can significantly reduce the complexity of user scheduling. In particular, for user scheduling strategies are proposed to improve performance include: (i) Select the highest-priority user, and try the next ν users for pairing, and perform precoding to check throughput. (ii) Select the highest-priority user, try the next ν users for pairing, and select the user with the highest orthogonality to the scheduled users and check it through precoding method. (iii) Select the highest-priority user, try the next ν users for pairing, and select the user with the highest orthogonality to the scheduled users and check threshold. (iv) Select the highest-priority user, and try the all users for pairing, and select the user with the highest fairness, and check throughput. Numerical results showed that the strategy-3 better than other strategies. In addition, we also propose the user grouping with priority method, numerical results show that indeed can achieve better fairness.