Research On Technologies Of Radio Resource Management For 5G

In recent years,mobile communication technology has continued to develop rapidly.Although radio resource management has existed for a long time,as an important part of mobile communication,it also needs to adapt to the new technologies and needs.Now the mobile communication system is in the period of transition from the fourth generation(4G)to the fifth generation(5G).More novel technologies and communication scenarios are emerging.Thus the traditional methods of radio resource management need to make more changes to adapt it.For example,multiple-input multiple-output(MIMO)systems have a trend toward massive MIMO.User grouping in radio resource management needs to consider this impact.In addition,the introduction of wireless network virtualization requires methods for user slicing and scheduling to meet the needs of different virtual networks(VNs).Meanwhile,with the rise of ultra-dense networks(UDNs),the demand for reducing inter-cell interference has further increased.Thus the inter-cell interference management is still worth studying.Based on these discussions,this thesis studies the radio resource management technologies for 5G,aiming to maintain the effectiveness and vitality of radio resource management.The main research contents of this thesis are as follows:(1)For massive MIMO,the user grouping and scheduling techniques in multi-user MIMO systems are mainly studied.In the single-antenna user scenarios,the impact of the number of user groups on system performance is first studied,and the grouping and scheduling scheme in which the number of user groups can be dynamically adjusted is proposed.The scheme is mainly divided into two parts:user grouping and group number adjustment.The former completes the specific user grouping and scheduling,while the latter can dynamically adjust the number of groups according to the performance of the users in the group.Simulation results show that compared with the methods with fixed number of groups,the proposed method effectively improves the stability of the system and reduces the performance loss caused by the change in the number of group.In addition,user selection algorithms based on coefficient of correlation are commonly used in single-antenna user scenarios.The principle is to divide users with large correlation values into different groups to reduce performance loss.However,these algorithms do not give an accurate relationship between the coefficient of correlation and performance loss.To solve this problem,this thesis derives the exact relationship between the loss of signal-to-interference-plus-noise ratio(SINR)and coefficient of correlation in single-antenna user scenarios.Furthermore,according to this relationship,a better user selection method based on coefficient of correlation is proposed in massive MIMO systems.This algorithm reduces the average performance loss caused by the correlation of users in a group.In addition,this thesis also studies the multi-antenna user scenarios.In these scenarios,the user selection algorithm based on chordal distance is generally considered to be the scheme with better performance and lowest complexity.Based on the coefficient of correlation and matrix operation,a better user selection method is proposed in this thesis.Compared with the chordal distance method,the proposed scheme achieves higher throughput and lower complexity,which is of great significance to further reduce the running time of user selection algorithm in massive MIMO systems.(2)For the radio resource management in wireless network virtualization,a dynamic slicing and scheduling scheme is proposed from the perspective of the radio access network.The solution can be divided into two main parts:the user scheduling part and the resource adjustment part.The user scheduling part is used to schedule users in VNs with given number of resources.Users in different VNs can use different scheduling algorithms.In the resource adjustment part,according to the requirements of each VN and the performance of users,the system will adjust the resources of each VN to achieve better performance.This scheme provides a new idea of wireless network slicing,and refines the specific scheduling method after slicing,achieving better performance and isolation.Simulation results show that the proposed scheme can better meet the needs of users in different VNs.(3)With the increase of the number of small cells,inter-cell interference coordination(ICIC)has become a key issue.For the scenarios of heterogeneous network,this thesis proposes a centralized macrocell muting scheme to reduce cross-layer interference between cells.The proposed scheme improves the performance of users in small cells by muting the macrocell at the subchannel level,and at the same time takes into account the throughput and fairness between the cross-layer and same-layer users.In the solution,in a centralized manner,the macrocell will comprehensively consider factors such as the fairness and throughput of users in each cell,and determine whether muting is required on each subchannel.While the judgment is completed,the system also determines the users that need to be scheduled on the subchannel.Through this method,the base station can utilize time-frequency resources more efficiently,and can further reduce the effects of interference on system performance.Simulation results show that the proposed scheme improves the throughput and fairness of users efficiently.