Research On Throughput And Fairness Of Multi-user NOMA-IRS Systems

With the rapid development of mobile communication technology,the volume of mobile Internet services has shown a spurt of growth.Intelligent reflection surface(IRS),a highly scalable and low-cost wireless physical layer technology,is considered as one of the key technologies for green and cost-effective next-generation mobile communications.This technology can reconfigure the wireless channel state and improve the received signal-tonoise ratio at the user side by regulating the signal amplitude and phase on each reflection unit.To solve the problem of spectrum scarcity in IRS system,non-orthogonal multiple access(NOMA)technology is introduced to build NOMA-IRS systems,which can improve the spectrum utilization and the multi-user access capability by superimposing multiple signals in the power domain.In obstacle-dense urban environments,low channel gain is often caused by blocked directthrough links.The introduction of NOMA-IRS system forms an additional reflected link that can effectively improve the link quality and enhance the system capacity.Therefore,in this thesis,the throughput maximization problem of a multi-user NOMA-IRS system is considered under the minimum transmission rate constraint.Since IRS not only introduces an additional modal constraint,but also makes the reflected beam vector and the power allocation coefficient highly coupled at the received signal to interference plus noise ratio(SINR).This leads to the non-convexity and high nonlinearity of the proposed problem.Therefore,this thesis proposes a joint optimization scheme of power allocation at the base station and reflection beamforming at the IRS.Firstly,a closed-form solution of the optimal power allocation scheme for users is derived based on the optimal decoding order of successive interference cancellation.Subsequently,a reflection beamforming optimization algorithm based on semi-definite relaxation(SDR)and arithmetic-geometric mean(AGM)is proposed with a fixed power allocation scheme,which gradually improves the system throughput by alternately optimizing the transmit power at the base station and the reflection beamforming vector at the IRS.In the simulation analysis,this thesis investigates the variation of system throughput achieved by the proposed AGM-SDR alternating optimization scheme with the transmit power of the base station,the number of IRS arrays,and the number of users.It is demonstrated that the proposed scheme significantly outperforms the random IRS,frequency division multiple access,and space division multiple access schemes.It also further elaborates the phenomenon of uneven resource allocation that occurs in the throughput maximization system.In multi-user systems,excessive pursuit of throughput maximization can lead to large resource occupation by individual users,resulting in significant rate differences among users and significantly affecting system fairness.Therefore,this thesis further investigates the fairness problem of multi-user NOMA-IRS system,and tries to maximize the minimum received SINR among users in a practical scenario where the RF links are limited.To solve this problem,we first cluster the users according to the number of RF links,use a joint "transmit-reflect" beamforming scheme to maximize the minimum channel capacity of each cluster,and finally allocate the power within the cluster to maximize the minimum channel capacity of users within the cluster.In this thesis,a maximum-minimum SINR fractional programming problem is constructed to implement the joint "transmit-reflect" beamforming optimization.Therefore,an AGM-SDR algorithm is proposed to maximize the minimum channel capacity in each cluster by alternately optimizing the transmit beamforming at the base station side and the reflect beamforming at the IRS side in each iteration.In addition,this thesis proposes a user clustering scheme based on equivalent channel correlation based on the number of RF chains and the combined transmit-reflective channel to enhance the user minimum channel capacity by exploiting the beam multiplexing gain.Finally,a binary search algorithm is used to solve the intra-cluster power allocation problem to maximize the minimum channel capacity among users.The complexity analysis and simulation results show that this scheme is able to obtain a minimum SINR improvement of 71% with polynomial level complexity compared to the zero-forcing scheme.In contrast to the maximum ratio transmission scheme,the user’s minimum SINR increases continuously with the increasing transmit power at the base station.