Research On Transmission Scheme For Reconfigurable Intelligent Surface Assisted Uplink NOMA System

According to the development law of "exploiting one generation and exploring one generation" for mobile communications,the research on B5G/6G has been rolled out all over the world.To maintain that China is at the leading edge in 6G,the Ministry of Science and Technology and other departments jointly launched the China 6G R&D project in November 2019.At present,many R&D institutions at home and abroad treat reconfigurable intelligent surface(RIS)as one of the potential technologies of 6G;In addition,due to the limitation of wireless resources,it is improssible to meet the requirements of huge traffic and massive connections by using traditional orthogonal multiple access in the future.Since the number of concurrent connections supported by non-orthogonal multiple access(NOMA)technology is not limited by the number of resource blocks,the NOMA has been received much attention from the academia and industry communities.The deep integration of RIS and NOMA is advocated as an effective solution to meet the demand of huge traffic and massive connection in future communication systems.In this thesis,aiming at the shortcomings of the existing research methods in power efficiency,on the premise of satisfying the differentiated service quality of users,a complete RIS-assisted NOMA system uplink low-power scheme is designed,which can substantially reduce the energy consumption of the future communication system and provide technical support to achieve the target of "carbon neutrality and carbon peak" for the information and communication industry.Firstly,we study the RIS-assisted uplink NOMA system,and propose a problem of minimizing the total transmission power of the system on the premise of satisfying the quality of service(Qo S)of each user.We first use the contradiction theory to derive the relationship between the optimal transmission power of users and the phase-shift matrix variables of RIS,and thus turn the original optimization problem into a pure phase shift optimization problem.To solve the transformed problem,we propose two different methods of phase shift design algorithms,namely the semi-positive definite relaxation method and the block coordinate descent method.The simulation results show the total transmission power of the system versus the Qo S of users,the number of reconfigurable units of RIS,and the number of users.It is shown that the proposed RIS-aided NOMA scheme is superior to several baselines in terms of total transmit power.Secondly,we explore the uplink NOMA system assisted by double cooperative RISs,and propose a problem of minimizing the total transmission power of the system on the premise of satisfying the Qo S of each user.According to the deployment position of BS and two RISs,the users are divided into two groups.In this system,there are dual reflection links assisted by two RIS.In order to deal with the non-convex problem with coupled variables,we first drive the relationship between the transmission power of users in two user groups and the phase shift variables of two RISs.Then,we propose an alternating optimization(AO)method to solve this phase shift optimization problem.The relationship between the total transmission power of the system,user Qo S and the reconfigurable intelligent unit of RIS is analyzed by simulations.On the premise of fixing the total reconfigurable units of two RISs,the optimal scheme with the lowest power consumption is used to allocate the unit numbers between two RISs.In addition,the convergence of the proposed algorithm is demonstrated.Finally,we consider the uplink NOMA system assisted by simultaneous transmitting and reflecting(STAR)RIS,and formulate an optimization problem of minimizing the total transmission power on the premise of satisfying the Qo S of each user.According to the deployment positions of the STAR-RIS and the BS,all users covered by the BS are divided into three categories,and the original optimization problem is further divided into three sub-problems according to the corresponding relationship between the constraints and the optimization variables.We further develop an AO method to address these sub-problems.The simulation results show the total transmission power of the system versus the Qo S of users and the number of reconfigurable units of STAR-RIS.Compared with the refection-only RIS which has the function of reflecting signals,the system assisted by STAR-RIS can bring additional performance gain.