Research On Wireless Communication System Based On Non-Orthogonal Multiple Access And Network Slicing Technology

In recent years,with the continuous growth of mobile data services and the diversification of service types and service demands,non-orthogonal multiple access(NOMA),multiple-input multiple-output(MIMO),network slicing and other the fifth generation(5G)key technologies have developed rapidly.In addition,user clustering and radio resource scheduling have been hot topic of research in the field of wireless communication.Especially in the period of transition from the fourth generation(4G)to the 5G.More and more new technologies and new service environments are appearing,and the traditional user clustering and wireless resource scheduling methods have to be adjusted and changed.For example,with the increase in the amount of base stations and equipment density,multi-cell systems have emerged,and user clustering has to consider the impact of these factors.Moreover,in order to meet the communication needs of multiple service types,network slicing techniques have been used to slice the wireless resources,which also requires consideration of the resource scheduling based on network slicing.In this background,this paper has researched the user clustering algorithm of the MIMO-NOMA multi-cell system and the resource scheduling problem for network slicing.The main research of this thesis are as follows:Firstly,the MIMO-NOMA multi-cell downlink system model is constructed,and the user clustering algorithm of this system is researched.Inter-cell interference is introduced,and the channel model of users and signal-to-interference-plus-noise ratio(SINR)relational equation are further derived.From the perspective of ensuring user fairness and system throughput,the user clustering algorithm based on channel gain difference and channel correlation is proposed,and the central transition matrix is constructed by combining the channel gain difference and channel correlation.Combined with the improved Kuhn-Munkres algorithm,the central transition matrix is used as an input of this algorithm to obtain the current user matching results,and then the user set is updated to readjust the central transition matrix.The capacity is compared to all matching results and the best user matching result is output.Compared with the classical algorithms of the same type,the simulation results show that the proposed algorithm can effectively improve the system throughput and verifies the feasibility of the algorithm.In addition,this thesis researches the dynamic resource allocation algorithm for network slicing based on the two-tier architecture model.Firstly,the quality of experience(QoE)of users is discussed in the user device layer and the slicing priority of users is researched in the slice control layer.Then,from the perspective of slices and intra-slice users,a two-order priority network slice resource allocation algorithm based on particle swarm optimization algorithm is proposed in order to guarantee the fairness of users and improve the system throughput.By first determining the priority of slices and submitting the current stable slice resource allocation scheme,and then making the secondary allocation of resources according to the priority of users within slices.The system resource base,node information of slices and users are updated and the new resource allocation scheme is reacquired in each loop iteration until all slices and users have completed resource allocation.Finally,the system performance is simulated and compared with the traditional resource allocation algorithm based on network slicing.The simulation results prove that the proposed algorithm can guarantee the user satisfaction and improve the system performance,which verifies the effectiveness of the algorithm.