Research On Key Technologies Of MIMO-NOMA System

To improve the energy efficiency and ability of massive connections of mobile communication systems,it’s necessary to investigate efficient Non-Orthogonal Multiple Access(NOMA)schemes and combine them with Multiple-Input Multiple-Output(MIMO),millimeter wave(mm Wave)and Intelligent Reflecting Surface(IRS)technologies.This can provide higher data transmission rates and lower power consumption for future wireless networks.Based on this,this thesis studies user clustering and power allocation problems in mm Wave MIMO-NOMA systems and energy optimization problems in dual IRS-assisted MIMO-NOMA systems.The specific work is as follows:(1)For millimeter wave MIMO-NOMA system: To reduce the computational complexity and improve the system energy efficiency,an improved user clustering scheme based on cluster-head selection is firstly designed.The threshold and the number of clusters are determined dynamically according to the real channel,so that the clustering results are more suitable for the actual situation.Then power allocation is designed with the goal of maximizing the weighted sum of spectral efficiency and energy efficiency of the system and solved by an improved meta-heuristic algorithm.By introducing new vector components and adding cosine perturbation to Particle Swarm Optimization algorithm,the algorithm converges to the global optimal value more quickly.Sand Cat Swarm Optimization algorithm is integrated to make the algorithm more accurate.The simulation results show that compared with the existing algorithms,the spectral and energy efficiency of the proposed scheme are better than the traditional schemes,and it is more suitable for multi-user cases.(2)For IRS-assisted wireless communication systems: This thesis studies double-IRS-assisted NOMA scenarios.To avoid excessive computational complexity,a two-stage optimization algorithm is used to decompose the original problem into two sub-problems.The first sub-problem is about reflection beamforming matrix variables at IRSs;the second is about active beamforming matrix variables at base station side.For the first sub-problem: a nested Riemannian manifold optimization algorithm is proposed which simultaneously optimizes reflection coefficients at two IRSs during each iteration process to maximize the sum of strong users’ equivalent channel gain within all user clusters.For the second sub-problem: to minimize total transmission power under the premise of ensuring the quality of communication,QR decomposition is used to eliminate the inter-cluster interference and reduce problem size,then semi-definite relaxation is used to make each transmission energy minimization problem convex.Simulation results show that the proposed scheme can achieve good performance with low computational complexity,and the proposed algorithm can be simply extended to more IRS assisted cases.