Research On User Scheduling And Precoding For Large-scale Antenna Systems

Since it can effectively improve the spectral efficiency and energy efficiency,explore the space resources,and greatly speed up the transmission rate,large-scale antenna technique is one of the key techniques in the fifth generation mobile communication(5G).Whilst,as the key of Internet of Things(Io T)and the great improvement of spectral efficiency(SE)in 5G,the large-scale antenna technique also faces the severe challenges: Firstly,compared with the great increase of the number of users,the hardware resources,such as the radio frequency(RF)chains and the assistant relays,are quite limited.How could we maximize the system performance with the given hardware resources,and how to save the uplink feedback resources? Secondly,in the dense and distributed networks,there are multiple large-scale antenna transmitting nodes.Thus,each user suffers from the multi-point interference from multiple large-scale antenna transmitters.How could we handle the multi-point interference via the precoding technique? Thirdly,the high-directional beam in large-scale antenna centralized millimeter wave systems contributes to the great directional beam gain,but the directional beam results in the fatal problem,that is,the directional gain is terribly easy to be affected by the state of user,such as the movement,blockage and movement+blockage,etc.How could we ensure the stable communication of the mobile users in large-scale antenna millimeter wave systems? The large-scale antenna,multi-point transmission,millimeter wave can greatly increase the efficiency and transmission rate,but it is still needed to ensure the system stability.The thesis starts from the different centralized and distributed large-scale antenna scenarios,focuses on user scheduling and precoding techniques,and tries to improve the system spectral efficiency and stability.The thesis utilizes the user scheduling technique to select the appropriate users for improving the system performance,and utilizes the precoding technique to decrease the inter-user interference and improve the stability.Through jointly focusing on the user scheduling and precoding techniques,we can ensure the stable improvement of performance in different large-scale antenna systems.The main contributions include:1.Focusing on the huge feedback problem due to the large-scale antenna,the thesis proposes the cooperative user scheduling algorithm in centralized large-scale antenna systems.The algorithm can effectively reduce the feedback overload in the uplink channel with a slight degradation of the system sum rate performance.Thus,it can save the spectral resources and improve the SE.In large-scale antenna systems,the numbers of users and transmitting antennas increase greatly,which leads to the increase of the load in the uplink channel.When BS acquires the exact channel state information(CSI),the channel information would occupy the numerous bandwidth resources,which decreases the SE.The proposed algorithm filters out the users with the bad channel condition via the device to device(D2D)property among users.After the filter,the allowed users feed back their channel information to BS.Then,the BS,according to the fed state information,selects the final user set for data transmission based on the semi-orthogonal user scheduling algorithm.According to the comprehensive theoretical analysis and the simulations,the proposed algorithm can effectively decrease the load of the uplink channel and maintain the system sum rate performance.2.Focusing on the problem that the traditional central user scheduling algorithms just consider the strength of useful signals in the relay-assisting large-scale antenna system,the thesis proposes the pricing user scheduling algorithm in relay-assisting large-scale antenna systems.The relays are utilized for “reducing” the distances between the BS and the far users to relieve the attenuation of transmitting power and improve the signal to interference-plus-noise ratio(SINR).Considering there are multiple assistant relays in the cell,and each relay would interfere other transmission pairs,the pricing user scheduling algorithm is proposed,which takes the strength of interference on other transmission pairs as the price so that it schedules the users with the better channel state and less interference on others.Except that,the low complexity equal gain transmission(EGT)precoding algorithm is proposed for distributed transmitters,and the theoretical upper bounds of sum rates in the first and second hops are given.The section suppresses the multi-point interference via the joint research on user scheduling and precoding.Based on the theoretical analysis and simulations,the proposed scheduling algorithm can effectively improve the sum rate performance in the relay-assisting large-scale antenna system.3.Focusing on the problem of the high computational complexity of the traditional iterative interference alignment algorithms,the thesis respectively proposes the interference alignment(IA)algorithms for the one-to-one and one-to-many interference channels,which is based on the multi-point joint precoding.In dense and distributed networks,there are multiple transmitters simultaneously,thus each user suffers from the multi-point interference from multiple large-scale antenna transmitters.The thesis,based on the nature of the multi-point interference – interference channel,proposes the low complexity interference alignment algorithms to decrease the computational complexity and computational time delay.Different from the traditional IA algorithms,the proposed IA algorithm fixes the precoding matrices of partial transmitters and directly aligns the interference subspace from different transmitters,which constructs the linear homogeneous equations of the precoding matrices instead of iteratively altering the directions of the precoding and combining matrices.Thus,the proposed IA algorithm can greatly decrease the computational complexity and maintain the stable communication,which is also proved via the complexity analysis and simulations.4.Focusing on the problem of the stable communication of the mobile users in centralized large-scale antenna millimeter wave systems,the thesis proposes the adaptive hybrid precoding algorithm for the mobile users.The sparse channel and highly directional beams in large-scale antenna millimeter wave systems would greatly affect the beam gain of mobile users.For effectively compensating the performance attenuation due to the mobility of mobile users,the thesis,firstly,proposes the state detection algorithm based on the channel state information.Through comparing the fed channel state information at different time slots,the states of different mobile users can be obtained.Then,based on the different states,the thesis respectively proposes the beam combining and adaptive hybrid precoding algorithms to compensate the performance attenuation.Based on the simulations,the adaptive hybrid precoding algorithm can effectively relieve the attenuation,maintain the sum rate performance and extend the beam training period to improve the stability and time-frequency efficiency.The research in the thesis can enrich the theories of user scheduling,precoding and interference alignment algorithms in different centralized and distributed large-scale antenna scenarios,improve the transmission performance and system stability.The thesis can lay the foundation for the multi-point transmission and high-frequency transmission,and provide the theoretical basis for the relative transmission schemes.